Re: p values
This is a multi-part message in MIME format. --5C80BC06D9EDF21501161EC3 Content-Type: multipart/alternative; boundary="30A9826BA49F60D5C9B364CC" --30A9826BA49F60D5C9B364CC Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit The word in your comment which catches my eye is "accurately" in >are we able to accurately say ... that if we had the results as in TEST C >... that we are more convinced that the null of 100 is NOT true ... than >for TEST B or TEST A? The question of "being convinced" is clearly a subjective one, what convinces you may not convince me, naturally. How we apply "accurately" to it, I don't know. If one takes the position advocated by L. J. Savage, that one wishes to avoid inconsistency in his judgements, then the whole of Bayesian probability theory is available as a normative theory of how we "should" make judgements about evidence. Significance tests have been shown to violate this consistency requirement in a number of ways, but that doesn't mean that they can't be used as an index of the strength of evidence, everything else being equal. >no ifs, ands, or buts or it depends on this or that ... just straight >talking ... can we say this? Of course, if you want a simple answer without conditions, you have to ask a simple question. I can't imagine anyone giving you a "straight talking" answer without qualifications to this question. --30A9826BA49F60D5C9B364CC Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit The word in your comment which catches my eye is "accurately" in >are we able to accurately say ... that if we had the results as in TEST C >... that we are more convinced that the null of 100 is NOT true ... than >for TEST B or TEST A? The question of "being convinced" is clearly a subjective one, what convinces you may not convince me, naturally. How we apply "accurately" to it, I don't know. If one takes the position advocated by L. J. Savage, that one wishes to avoid inconsistency in his judgements, then the whole of Bayesian probability theory is available as a normative theory of how we "should" make judgements about evidence. Significance tests have been shown to violate this consistency requirement in a number of ways, but that doesn't mean that they can't be used as an index of the strength of evidence, everything else being equal. >no ifs, ands, or buts or it depends on this or that ... just straight >talking ... can we say this? Of course, if you want a simple answer without conditions, you have to ask a simple question. I can't imagine anyone giving you a "straight talking" answer without qualifications to this question. --30A9826BA49F60D5C9B364CC-- --5C80BC06D9EDF21501161EC3 Content-Type: text/x-vcard; charset=us-ascii; name="rabeldin.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Richard A. Beldin Content-Disposition: attachment; filename="rabeldin.vcf" begin:vcard n:Beldin;Richard tel;home:787-255-2142 x-mozilla-html:TRUE url:netdial.caribe.net/~rabeldin/Home.html org:BELDIN Consulting Services version:2.1 email;internet:[EMAIL PROTECTED] title:Professional Statistician (retired) adr;quoted-printable:;;PO Box 716=0D=0A;Boquerón;PR;00622; fn:Richard A. Beldin end:vcard --5C80BC06D9EDF21501161EC3-- = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: p values
This is a multi-part message in MIME format. --152074D07C4F46BEF5496A61 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit One should note that this comment was taken from an article critical of significance testing. I understand Sterne and Smith's comment to characterize a point of view which they do not support, but which they project onto consumers of statistics. (I think they could have been a bit clearer about their own point of view, but this is how I read the statement in the context of what follows.) Generally, statisticians with a background in theory of inference will agree with the general thrust of the article: Significance tests are often performed and reported and interpreted in a mechanical way which may be misleading. The suggested guidelines (which suggest against use of the phrase "statistically significant" without further qualification) will generate considerable support among those who have studied the theories of frequentist and Bayesian inference. The only support for the significance test is in the use of the P-level as an index to the strength of the evidence. Personally, I would qualify that with the standard "Other things being equal...". --152074D07C4F46BEF5496A61 Content-Type: text/x-vcard; charset=us-ascii; name="rabeldin.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Richard A. Beldin Content-Disposition: attachment; filename="rabeldin.vcf" begin:vcard n:Beldin;Richard tel;home:787-255-2142 x-mozilla-html:TRUE url:netdial.caribe.net/~rabeldin/Home.html org:BELDIN Consulting Services version:2.1 email;internet:[EMAIL PROTECTED] title:Professional Statistician (retired) adr;quoted-printable:;;PO Box 716=0D=0A;Boquerón;PR;00622; fn:Richard A. Beldin end:vcard --152074D07C4F46BEF5496A61-- = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Help Log-Probit in Excel
This is a multi-part message in MIME format. --46D0737844C9CF0BEDE1D80F Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit I don't know where that came from. I studied from Finney as well. I guess I'm suffering from old-timer's disease. :-) --46D0737844C9CF0BEDE1D80F Content-Type: text/x-vcard; charset=us-ascii; name="rabeldin.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Richard A. Beldin Content-Disposition: attachment; filename="rabeldin.vcf" begin:vcard n:Beldin;Richard tel;home:787-255-2142 x-mozilla-html:TRUE url:netdial.caribe.net/~rabeldin/Home.html org:BELDIN Consulting Services version:2.1 email;internet:[EMAIL PROTECTED] title:Professional Statistician (retired) adr;quoted-printable:;;PO Box 716=0D=0A;Boquerón;PR;00622; fn:Richard A. Beldin end:vcard --46D0737844C9CF0BEDE1D80F-- = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Help Log-Probit in Excel
This is a multi-part message in MIME format. --ECCA490F99BC503A5BC927C1 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit The transform from Z scores to probits uses a constant 6 translation, not 5. I don't know if that solves your problem, but it might eliminate some confusion. --ECCA490F99BC503A5BC927C1 Content-Type: text/x-vcard; charset=us-ascii; name="rabeldin.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Richard A. Beldin Content-Disposition: attachment; filename="rabeldin.vcf" begin:vcard n:Beldin;Richard tel;home:787-255-2142 x-mozilla-html:TRUE url:netdial.caribe.net/~rabeldin/Home.html org:BELDIN Consulting Services version:2.1 email;internet:[EMAIL PROTECTED] title:Professional Statistician (retired) adr;quoted-printable:;;PO Box 716=0D=0A;Boquerón;PR;00622; fn:Richard A. Beldin end:vcard --ECCA490F99BC503A5BC927C1-- = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Excel Histograms
This is a multi-part message in MIME format. --015C2EDE4247EED2D42E7582 Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit The easiest way I know is to make a frequency table and tell Excel to turn it into a bar graph. You can set the class boundaries to be whatever you like. Your table should look like: From To Count 0.00 0.99 =COUNT(data,AND(">RC[-2]","<=RC[-1]")) etc... --015C2EDE4247EED2D42E7582 Content-Type: text/x-vcard; charset=us-ascii; name="rabeldin.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Richard A. Beldin Content-Disposition: attachment; filename="rabeldin.vcf" begin:vcard n:Beldin;Richard tel;home:787-255-2142 x-mozilla-html:TRUE url:netdial.caribe.net/~rabeldin/Home.html org:BELDIN Consulting Services version:2.1 email;internet:[EMAIL PROTECTED] title:Professional Statistician (retired) adr;quoted-printable:;;PO Box 716=0D=0A;Boquerón;PR;00622; fn:Richard A. Beldin end:vcard --015C2EDE4247EED2D42E7582-- = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: I dont know what the question is asking me? Please HELP!
This is a multi-part message in MIME format. --82BC8D92CC03F9E9B6B85561 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit In this context, "substantiate" is used to mean "provide evidence for". Its a stretch from "value" but is commonly used by people who like to inflate their syllable count. :-) --82BC8D92CC03F9E9B6B85561 Content-Type: text/x-vcard; charset=us-ascii; name="rabeldin.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Richard A. Beldin Content-Disposition: attachment; filename="rabeldin.vcf" begin:vcard n:Beldin;Richard tel;home:787-255-2142 x-mozilla-html:TRUE url:netdial.caribe.net/~rabeldin/Home.html org:BELDIN Consulting Services version:2.1 email;internet:[EMAIL PROTECTED] title:Professional Statistician (retired) adr;quoted-printable:;;PO Box 716=0D=0A;Boquerón;PR;00622; fn:Richard A. Beldin end:vcard --82BC8D92CC03F9E9B6B85561-- = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: conditional probability
This is a multi-part message in MIME format. --BAA90E5A97D4B52B78652CCA Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Terry, There is no "canonical" probability model for such a situation. First one has to specify his probability model and then one can calculate probabilities for a variety of events. In your first question, "if you observe the person for 15 minutes" could be interpreted as "a randomly selected period of 15 minutes". In that case, one could calculate the probability one divided by the number of consecutive 15 minute periods in the 7 hours. Depending on exactly how one constrains that 15 minute period, one could get answers as different as 1/(7*4) and 1/(7*60). Building probability models doesn't generate information out of nothing, it transforms it from one form to another. Dick --BAA90E5A97D4B52B78652CCA Content-Type: text/x-vcard; charset=us-ascii; name="rabeldin.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Richard A. Beldin Content-Disposition: attachment; filename="rabeldin.vcf" begin:vcard n:Beldin;Richard tel;home:787-255-2142 x-mozilla-html:TRUE url:netdial.caribe.net/~rabeldin/Home.html org:BELDIN Consulting Services version:2.1 email;internet:[EMAIL PROTECTED] title:Professional Statistician (retired) adr;quoted-printable:;;PO Box 716=0D=0A;Boquerón;PR;00622; fn:Richard A. Beldin end:vcard --BAA90E5A97D4B52B78652CCA-- = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Problem on the probability of death
This is a multi-part message in MIME format. --6868FFBD26298125D6F02FAE Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit A common approach is to change the focus from the inevitable fact of death to the duration of life within the treatment and control groups. Presumably, if the treatment is effective, the treated subjects will survive longer than the controls. --6868FFBD26298125D6F02FAE Content-Type: text/x-vcard; charset=us-ascii; name="rabeldin.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Richard A. Beldin Content-Disposition: attachment; filename="rabeldin.vcf" begin:vcard n:Beldin;Richard tel;home:787-255-2142 x-mozilla-html:TRUE url:netdial.caribe.net/~rabeldin/Home.html org:BELDIN Consulting Services version:2.1 email;internet:[EMAIL PROTECTED] title:Professional Statistician (retired) adr;quoted-printable:;;PO Box 716=0D=0A;Boquerón;PR;00622; fn:Richard A. Beldin end:vcard --6868FFBD26298125D6F02FAE-- = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: MA MCAS statistical fallacy
This is a multi-part message in MIME format. --A0316B6769484E5B80B17052 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit In one way or another, we have to give the slower students more time. We can do it by making courses which allow students to progress (or not) at their own pace or by flunking them so they can do it all over. The former is more efficient, but the latter works too. What doesn't work is pretending that slower learners completed their course in the same time as the quicker ones. --A0316B6769484E5B80B17052 Content-Type: text/x-vcard; charset=us-ascii; name="rabeldin.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Richard A. Beldin Content-Disposition: attachment; filename="rabeldin.vcf" begin:vcard n:Beldin;Richard tel;home:787-255-2142 x-mozilla-html:TRUE url:netdial.caribe.net/~rabeldin/Home.html org:BELDIN Consulting Services version:2.1 email;internet:[EMAIL PROTECTED] title:Professional Statistician (retired) adr;quoted-printable:;;PO Box 716=0D=0A;Boquerón;PR;00622; fn:Richard A. Beldin end:vcard --A0316B6769484E5B80B17052-- = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Scale on survey questionnaire
This is a multi-part message in MIME format. --7BA592567A43C83B890B0B6C Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit The ? is guaranteed to be a catch all. Understandable statistical analysis will be conditional on receiving other responses. Nobody should believe that you can build an ordinal scale out of such data. Ambiguity is essential in some situations. --7BA592567A43C83B890B0B6C Content-Type: text/x-vcard; charset=us-ascii; name="rabeldin.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Richard A. Beldin Content-Disposition: attachment; filename="rabeldin.vcf" begin:vcard n:Beldin;Richard tel;home:787-255-2142 x-mozilla-html:TRUE url:netdial.caribe.net/~rabeldin/Home.html org:BELDIN Consulting Services version:2.1 email;internet:[EMAIL PROTECTED] title:Professional Statistician (retired) adr;quoted-printable:;;PO Box 716=0D=0A;Boquerón;PR;00622; fn:Richard A. Beldin end:vcard --7BA592567A43C83B890B0B6C-- = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Scale on survey questionnaire
This is a multi-part message in MIME format. --D59B8CE81F610975488820BF Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit "For budget reasons it wasn't possible to translate the questionaire..." That seems to hide some interesting issues. You pay consultants to generate an instrument that many of the respondents can't deal with effectively? And you expect management to pay attention to the results of such weak research? I have worked as a manager and an internal consultant in multilingual industrial settings. Serious managers insist on effective communications with their employees for a number of good business reasons. I would hope that customer and supplier communications would be treated as seriously. Again, I repeat, the terminology used in the scales is not a "statistical" issue, but a communications issue. --D59B8CE81F610975488820BF Content-Type: text/x-vcard; charset=us-ascii; name="rabeldin.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Richard A. Beldin Content-Disposition: attachment; filename="rabeldin.vcf" begin:vcard n:Beldin;Richard tel;home:787-255-2142 x-mozilla-html:TRUE url:netdial.caribe.net/~rabeldin/Home.html org:BELDIN Consulting Services version:2.1 email;internet:[EMAIL PROTECTED] title:Professional Statistician (retired) adr;quoted-printable:;;PO Box 716=0D=0A;Boquerón;PR;00622; fn:Richard A. Beldin end:vcard --D59B8CE81F610975488820BF-- = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Scale on survey questionnaire
This is a multi-part message in MIME format. --3937A5B7BBB144812A431A4E Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit The question mark is just being used to avoid using words like "no opinion" or "neutral", presumably assuming that most foks would make a reasonable interpretation. Actually, the terms used in a scale of responses is more a language comprehension issue than a statistics issue. As far as the question mark being ambiguous, I think that is intentional. "Neutral" and "No opinion" and "Don't care" all are rather ambiguous, but ? catches the sense well, IMO. Dick --3937A5B7BBB144812A431A4E Content-Type: text/x-vcard; charset=us-ascii; name="rabeldin.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Richard A. Beldin Content-Disposition: attachment; filename="rabeldin.vcf" begin:vcard n:Beldin;Richard tel;home:787-255-2142 x-mozilla-html:TRUE url:netdial.caribe.net/~rabeldin/Home.html org:BELDIN Consulting Services version:2.1 email;internet:[EMAIL PROTECTED] title:Professional Statistician (retired) adr;quoted-printable:;;PO Box 716=0D=0A;Boquerón;PR;00622; fn:Richard A. Beldin end:vcard --3937A5B7BBB144812A431A4E-- = Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ =
Re: Sex Ratio, part 2
Wolfgang: As I remember my course in demography many years ago, there were indications that the prenatal mortality was excessively higher in males than in females. An evolutionary argument would claim that whatever behavior equalizes (approximately) the birth ratio would be advantageous. The other scrap of data I have has to do with the pH of the vaginal tract during the normal female cycle. As the pH varies, there is a variable survival rate of spermatazoa with the different sex chromosomes. That goes along with the higher frequency of intercourse theory. Wolfgang Rolke wrote: > Thanks to all those who have answered my previous email. Here are a > couple of comments: > > 1) When I sent my inquiry I expected to get many replies of the form > "Look at this web site (journal article ..), it will answer all your > questions". Considering that the fact that slightly more boys than girls > are born has been known for such a long time (at least 300 years) one > would think that a great deal is known and that many studies have been > done. I also did a search on the web. Unfortunately any web search > involving the keyword "sex" returns a great many quite unrelated web > sites, and none of the useful ones was all that enlightening. > > 2) As for the reason for the observed inequality in birthrates, there > seem to be several: > > a) Environmental reasons such as pollution, climate, living > conditions ect. Considering how long this problem has been studied I > would have expected to see some study to show a correlation between any > such factor and the sex ratio. Of course if such a relationship exists > it should hold not just in one study but in any study with the relevant > data. > > b) Evolution. The argument here seems to be that there is an > evolutionary advantage to having more boys born than girls, maybe > because of the higher mortality rate of men. I don't find this argument > very convincing , mostly because one man can impregnate any number of > women, and so if anything I would expect this effect to work in the > opposite direction. Also, it would be amazing to see an evolutionary > effect to have as subtle an influence as is observed here. > > c) Genetic reasons involving the X and Y chromosomes. If this is > true it appears to me that with our current knowledge of Genetics > somebody should have been able to give a convincing (and testable) > explanation for this phenomena > > d) hlau wrote: > > ["...When couples have sex more often, > > fertilization is more likely to occur early in the menstrual cycle, > > which apparently increases the odds of male conception. > > Some observers believe this conjecture explains why the percentage > > of baby boys has usually increased after major wars."] > > Now this is my favorite. I seem to have known this almost forever, > but I have never been sure whether or not it is really true or whether > it is an urban legend. Are there any studies that show a relationship > between frequency of intercourse and the sex of the baby? Now here is a > data set that my students could get excited about! (Pun intended) > > Wolfgang === This list is open to everyone. Occasionally, people lacking respect for other members of the list send messages that are inappropriate or unrelated to the list's discussion topics. Please just delete the offensive email. For information concerning the list, please see the following web page: http://jse.stat.ncsu.edu/ ===
Re: Sex ratio
Wolfgang: The State of New York has published vital statistics in table form for 1995-1997. Here is a reference to one of the 1997 tables. http://health.state.ny.us/nysdoh/vs97/tab04.htm. Sounds like an interesting introduction. Dick Wolfgang Rolke wrote: > Hi, > > I am looking for information (journal article, web site ect.) on the > difference in birth rates for boys and girls. Specifically I would like > to find some real data and possible explanations for this phenomena. It > seems to me that this would make a great example for an introductory > stats course. > > Thanks > > Wolfgang === This list is open to everyone. Occasionally, people lacking respect for other members of the list send messages that are inappropriate or unrelated to the list's discussion topics. Please just delete the offensive email. For information concerning the list, please see the following web page: http://jse.stat.ncsu.edu/ ===
Re: Sample Size for an Audit
You have several dimensions here. First you have the posibililty of variability among the supervisors. With only 20, you may as well give them all the test because by chance you could leave out those supervisors who will cause you trouble later on. Next, and more difficult is the variation in knowledge. What are the most important aspects for the audit? Is conceptual knowledge more important than ability to express themselves before their subordinates? What about their level of committment to the program? Is lip-service good enough? Is a paper-and-pencil test the most appropriate? What about on-the-spot inconspicuous observation of each work group for a period of time? Perhaps the expert judgement of someone in the management chain is as valid as any test? If you take the task too seriously, you will run out of time. Perhaps questions based on the instructional materials used are sufficient. Perhaps not. [EMAIL PROTECTED] wrote: > We are going to do a quality system audit (like ISO 9000). How do I > choose the sample size for a particular group of people? Let's say > that there are 20 supervisors and I will audit their knowledge of SPC, > how many should I choose for the audit? > > Sent via Deja.com http://www.deja.com/ > Before you buy.
Re: statistical event definition?
Well, we wouldn't try to analyze apples and autos in the same data set. :-) On the other hand, the similarity is sort of like what one requires for an efficient tabular data base. Whatever the sample space of events is, it should consist of events that we can think about together comfortably. If it takes a paragraph to describe each event, statistical methods aren't likely to be useful. We need to be able to code the data into compact data sets like measurements on a sample of objects, or a series of measurements on a single object. Sorry, I don't know of any formal criteria. Usually this is only discussed when somebody violates some statistician's common sense. Then it may be discussed in a rather combative atmosphere but privately, rather than publicly. Muriel Strand wrote: > i gather that a collection of events which is analyzed with statistics > must have sufficient similarity (between each event) for the analysis to > be accurate/precise. how similar is sufficient? can anyone recommend > refs (preferably books) that discuss this issue, and provide guidelines > for assuring sufficient similarity? does this consideration affect the > appropriate choice of model? > > thanks in advance for sharing your wisdom & experience. > > -- > Any resemblance of any of the above opinions to anybody's official > position is completely coincidental. > > Muriel Strand, P.E. > Air Resources Engineer > CA Air Resources Board > 2020 L Street > Sacramento, CA 59814 > 916-324-9661 > 916-327-8524 (fax) > www.arb.ca.gov
Re: Conditional probability problem
The ambiguity is in the initial statement. "Given an offspring is affected" can mean 1) We know that the i'th offspring is affected and the statuses of the other m-1 are unknown 2) We know the status of all m offspring and exactly one is affected. 3) We know the status of all m offspring and x (>0) are affected. Use the most specific condition that you can confirm. James Cui wrote: > I have a problem of calculating a conditional probability, and want to know > what is the right formula. > > Suppose a family consists of two parents and m offsprings. Consider a > disease that affects both genders equally. Given an offspring is affected, > what is the probability that neither two parents nor other m-1 offsprings > are affected. Am I right to use the following formula ? > > Pr(F1=0; F2=1) / Pr(F2>0) > > where F1 is the number of affections in parents, and F2 is the number of > affections among offsprings. > > Thanks, > James.
Re: addition of truncated normal distribution
My intuition tells me that although the sum of X1 and Y1 will not be exactly a truncated normal, you will find that a truncated normal offers a good approximation if both means are far from zero where the truncated part is small. If you simulate this system, you will develop a better feeling for it. "H. J. Wang" wrote: > Hi, > > Suppose X, Y are independent random variables with normal distributions. > The means and variances are different. Assume X1 and Y1 are random > variables with the probability distributions f(X | X>=0) and g(Y|Y>=0), > respectively. That is, X1 and Y1 the non-negative truncations of X and > Y, respectively. Does anyone know whether in this case Z = X1 + Y1 is > still a truncated normal? Any reference on this? Thanks in advance!
Re: Occupation codes?
The US Department of Labor has some statistics that are presented by occupation code. Apart from the cultural bias towards a mid-twentieth century pattern of employment in the US, they are comprehensive. Kimon Spiliopoulos wrote: > Dear all, > > I would appreciate any help on the following: > > I am trying to find a reliable profession/ occupation coding. The system > used by the local (Greek) Statistical Service won't do because it is a > general and much detailed activity-based codification (like NACE > coding). > > Is anybody aware of any more or less standardised/ commonly used such > system? In case the reply depends on the use, it's for grouping bank > clients. > > Thanks in advance, > > Kimon Spiliopoulos > > P.S. Indicative level of detail from some list I found in the Web: > > . > BA - Banking > BB - Barber > BC - Butcher > BD - Builder (Construction) > BE - Broker/Investment > BI - Biologist > BK - Baker > BL - Blacksmith > BM - Boilermaker > BO - Botanist > BP - Bookkeeper > BR - Bricklayer (Mason) > BS - Bldg Services (Janitor, Maid) > BT - Bartender > BU - Business > BY - Buyer/Purchaser/Merchandise > BZ - Baby Sitter/Nanny > ..
Re: grades approximated by multinomial?
There is no approximation if the students' grades are independent. The multinomial distribution only assumes independent trials and constant value of the p[i] parameters which sum to 1. Of course, a different class may have different values of the p[i], but that distribution will also be multinomial. EAKIN MARK E wrote: > As an example in class, I was considering using the letter grade > distribution in class as an example of an approximate multinomial > distribution. How close do you feel this approximation is? > > Mark Eakin > Associate Professor > Information Systems and Management Sciences Department > University of Texas at Arlington > [EMAIL PROTECTED] or > [EMAIL PROTECTED]
Re: an open review process in Statistics journals?
I once wrote a "hostile review" because the author was bragging about his current work, lots of hyperbole about how it would "revolutionize" something. I wouldn't have minded if the author had known who wrote it, but it could have been uncomfortable had I met him/her later.
Re: y2k confound
That's how we get "self-fulfilling prophecies"! Actually, many companies are so stingy with their IT investment that some kind of "scare scenario" may have been necessary to get their attention. Seems a shame, but that's life in the corporate world. Managers don't really take their computers seriously until consultants tell them to! Bob Hayden wrote: > - Forwarded message from Paige Miller - > > > > > I'm wondering if those spending/earning the billions are congratulating > > themselves on so "few problems" (We fixed that just right!!!) or if the > > problems existed in the first place. Now, if we'd only had a control > > group. > > I read somewhere that a state government agency deliberately left three > computers unfixed for Y2K and they crashed immediately and were useless. > > - End of forwarded message from Paige Miller - > > Sounds pretty unlikely. I'm sure there are thousands of computers out > there that were not updated and I have not heard of massive crashes. > Also, I'm not aware of any y2k problems that kill a system -- they > usually just mess up dates. > > Of my three machines at home, the ancient 486 laptop showed multiple > problems on testing, so I let the test program "fix" it. The two 686s > showed only one problem -- they would not automatically roll over to > y2k but could be advanced manually. So I left those alone. As it > turned out, they DID advance themselves automatically. So now to the > statistical issue;-) if other testing software was similarly > pessimistic then the y2k problem may have been overestimated. > > Another tests and measurement issue -- I heard one report on a talk > show that one facility found all its computers reading 4 JA 1980 on > New Year's Day. A y2k bug? Not exactly. I noted that one of the > test programs I used left the system clock set wrong. How wrong? > Well, if you did not notice the problem, then on New Years's Day the > clock would have read 4 JA 1980. I suspect this problem was caused by > the test program, not by a y2k bug! > > _ > | | Robert W. Hayden > | | Department of Mathematics > / | Plymouth State College MSC#29 >| | Plymouth, New Hampshire 03264 USA >| * | Rural Route 1, Box 10 > /| Ashland, NH 03217-9702 > | ) (603) 968-9914 (home) > L_/ [EMAIL PROTECTED] > fax (603) 535-2943 (work)
Re: grading on the curve
My professors at Michigan were also fair. Probably some of the same ones as yours. However, my students have not always been fair - to themselves. Last term (my last term, ending in June), I had one student in precalculus who reported that he had failed the course four times (not unusual, there were a half a dozen in the same boat). This time, he changed his tactics. He decided to actually do the exercises! Guess what? He earned a B! One of my favorite success stories. It makes the point that like the old adage, "You can take the kid to school but you can't make him learn"! "J. Williams" wrote: > I wonder if all of us have experienced angst in assigning marks that can > affect the academic careers of our students. I always took it very seriously > and was concerned about the student who fell below some arbitrary cutoff I > ordained. I started teaching back in the '60s and have seen the rise of > "grade inflation" and extraneous pressures to "equalize." My thinking now is > that we tend to treat our students in the manner we were treated as undergrad > and graduate students. My professors back at Michigan in the "Olden Days" > were fair to me...more than fair. The least I can do is treat my own students > similarly. This thread's diversity of thinking attests to the differing > perceptions of what is fair and to what constituencies? What is right for > employers? What is correct for society? What's best for the individual > student? At the very least, grading is not taken lightly by the various > respondents. I'm not so old I can't remember the pressures on me to get "A's" > and to look goodand keep my ego boosted.
Re: grading on the curve
The only take home exam I ever felt comfortable with was in a business calculus course. I assigned each student a different polynomial of degree 3 or 4. Their assignment was to obtain the definite integral between a and b (given constants) by each of four methods with no more than 10% difference among the solutions. The methods were symbolic integration, graphical integation, Simpson, and trapezoidal numerical integration. There was only one eager beaver who actually solved someone else's problem as well as her own. :-) "T.-S. Lim" wrote: > In article <[EMAIL PROTECTED]>, [EMAIL PROTECTED] says... > > > >Dear Dr. Dallal, > > > >I'm interested in your statement about the 'take home portion of the exam' > >and what weight you place upon an exam that can easily be done with the joint > >help of other individuals. How do you know what portion of the 'take home > >exam' is really being done by the student, ONE HIS OR HER OWN, without > >outside help? > > I don't like take-home exams at all. It's very hard (almost impossible) to > ensure that each student does the exam alone. It's true even for graduate > students (I'm speaking from my own personal experiences). > > > > -- > Tjen-Sien Lim > [EMAIL PROTECTED] > www.Recursive-Partitioning.com > > Get your free Web-based email! http://recursive-partitioning.zzn.com
Re: adjusting marks
I also want to add a bit about my predjudices. In my seventeen years in industry, I rarely heard of anyone getting praise for "trying". The emphasis was on "results", even at the cost of some formal policies. However, in the twelve years I spent in academia, both before and after my industrial work, I have heard of getting rewarded for "effort". Somehow, I think there is a correlation. Teachers don't get rewarded for results, but for effort. Maybe that's why we consider rewarding students in the same way. Educational institutions have not come to grips with measuring the effectiveness of teachers. It's about time we did!
Re: adjusting marks
When my students asked me (as a class) to grade on a curve, I suggested the following alternative. "Place N chips in a can. Let them marked in the following way: 10%F, 20%D, 40%C, 20%B, 10%A. Let each student pick a chip and leave the class, certain of his/her grade." For some reason, nobody ever wanted to do that! :-) Generic wrote: > My wife wants to adjust marks for a course she is marking. Does someone have > a formula or something for using a bell curve to move them up or down? > > I have done this sort of thing about 15 years ago, but I can't remember any > of it! > > --
Re: correlation
They are (unless otherwise described) calculated as Pearson product-moment correlations. The distinctions are either in terms of the variables being correlated (1,2,3,4,5) or in the format (6) in which the correlations are presented. haytham siala wrote: > Could you please answer this question? Correlation are given in different > tables under the headings: > > 1) Zero-Order Correlation between variables > 2) Intervariable correlations > 3) orrelation Between factors > 4) Pearson Correlation between Variables > 5) Intercorrelation between variables > 6) correlation matrix > > Do these have the same meaning but are named differently in different > contexts?
Re: teaching statistical methods by rules?
The question is: If you don't teach by rules, what will you use? In the Elementary Statistics course I used to teach, I had several different objectives. 1) Students should develop arithmetic reliability. 2) Students should learn how they can be tricked by statistical wizardry. 3) Students should learn how to understand what is written by responsible statisticians. In the Applied Statistics course, my objectives were different. 1) Students should learn to read and evaluate the professional literature in their discipline. 2) Students should learn to apply the basic ideas of statistical description and inference: estimation, simple hypothesis testing, regression, etc. 3) Students should learn that their experience is limited and gain the confidence to consult a professional statistician. In the Mathematical Statistics course, my objectives were primarily mathematical, not statistical. Statistics was seen as a technology built on the foundations of the infinitesimal and finite difference calculus. Different strokes for different folks!
Re: counting seeds
Consider the mechanisms which reduce the number of live seeds. Hypothesis 1. Soil quality determines what proportion will germinate. Crowding reduces the share of the available nutrient resources per seed. Hypothesis 2. Predation by birds is more likely when the density is higher so that the seeds are more visible. Hypothesis 3. Competition with other plants (weeds) determines the survival of the target seeds. I see no problem with the descriptive measures proposed, but the experiment should be tailored to the kinds of alternative hypotheses entertained. "Richard M. Barton" wrote: > A biology student came to me with a data analysis situation that I wasn't sure how >to deal with. Sound advice would be appreciated. > > Scenario: > > Ben has a number of 1 meter square plots where he placed one or more seeds: >50 plots with 1 seed >10 plots with 25 seeds >10 plots with 50 seeds > > He replicated that design with four species of seeds. > > He visited the plots every day for a week to count the number of seeds remaining. > > So the questions of interest are: >a) Does density have an effect on seed survival? >b) Does species have an effect on survival? >c) What does the data look like over time? > > We considered modelling/analyzing the data in two ways (using SAS): > > 1) with seed as the unit of analysis, using Proc Lifetest to generate survival >curves. > The problem: for medium and high density plots, seeds would not seem to be > independent. > > 2) with plot as the unit of analysis, using GLM to get a mixed model, where > time is a repeated measure and density and species are between groups factors. > Problem: low density (1 seed) plots have a dichotomous outcome, so much > of the data is non-normal. > > Any suggestions? Thanks. > > rick barton > >
Re: Censured data, Optimum Accelerated Censored Life
We really need something more about the context where you found the phrase "accelerated tests". That is not a standard term in general. I could guess, but my guess might be wrong. Jean-Pierre wrote: > Hi, > > I'm a french student and I would like to have many > informations. > > What are Accelerated tests ? > What are censored data (Type I and II) ? > > Thanks for your answer. > > Jean-Pierre > > * Sent from AltaVista http://www.altavista.com Where you can also find related Web >Pages, Images, Audios, Videos, News, and Shopping. Smart is Beautiful
Re: Overidentification in structural eqns
--0FE2A80BBDB78FA4819CDE7F Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit I suggest that you need to have some criterion to use as a tie breaker. Can you identify any of the observed characteristics which might be a proxy for a "goodness-of-fit" criterion? Dick Kenneth Benoit wrote: > Dear Colleagues: > > Perhaps someone can help me with this problem. I am trying to solve for > a number of parameters in three equations which are linked through > composition of the data. each model yields different parameter > estimates when estimated alone since the parameters are overidentified. > I'd be happy for any advice on the problem! > > Apologies for cross-postings. > > Ken Benoit > > - > Kenneth Benoit http://benoit.tcd.ie > Department of Political Science mailto:[EMAIL PROTECTED] > Trinity College Tel: 353-1-608-2491 > Dublin 2, Ireland Fax: 353-1-677-0546 > > > Kenneth Benoit > [EMAIL PROTECTED] > Nov 26, 1999 > > Consider a system where: > Y1 = X0+ (1-g11)b1 X1 + g21b2X2 + g31b3X3 Y2 = X4 + g12b1 X1 + (1-g22)b2X2 + g32b3X3 Y3 = X5 + g13b1 X1 + g23b2X2 + (1-g33)b3X3 > > and: > > 1 = Y1 + Y2 + Y3, 0 ³ Yi ³ 1.0 "i > 1 = X0 + X4 + X5, 0 ³ Xi ³ 1.0 "i > 1 = g11 + g12 + g13, 0 ³ gij ³ 1.0 "i,j > 1 = g21 + g22 + g33 > 1 = g31 + g32 + g33 > > GOAL: To estimate g's and b's. Problems: overidentification; effects > of the data items and some of the parameters summing to 1 which I > still don't fully understand. > > Background: This is for a voting transition study in Italy, where the > b's represent a the proportion of voters following a rational > proximity model, and the g's represent the discrete probability > distribution according to which non-rational voters distribute their > votes to one of three electoral coalitions (corresponding to the Y's). > I have data for all of the Y's and X's, which are proportions. > > Possible ways to simplify: > >* Set gij's to constants before estimation. >* Set gij = g* " i,j. >* Set b3 = 1. >* Set b1 = b2. > > --- > File translated fromTEXby TTH,version 2.56. > On 26 Nov 1999, 14:00. --0FE2A80BBDB78FA4819CDE7F Content-Type: text/html; charset=us-ascii Content-Transfer-Encoding: 7bit I suggest that you need to have some criterion to use as a tie breaker. Can you identify any of the observed characteristics which might be a proxy for a "goodness-of-fit" criterion? Dick Kenneth Benoit wrote: Dear Colleagues: Perhaps someone can help me with this problem. I am trying to solve for a number of parameters in three equations which are linked through composition of the data. each model yields different parameter estimates when estimated alone since the parameters are overidentified. I'd be happy for any advice on the problem! Apologies for cross-postings. Ken Benoit - Kenneth Benoit http://benoit.tcd.ie">http://benoit.tcd.ie Department of Political Science mailto:[EMAIL PROTECTED]">mailto:[EMAIL PROTECTED] Trinity College Tel: 353-1-608-2491 Dublin 2, Ireland Fax: 353-1-677-0546 Kenneth Benoit [EMAIL PROTECTED] Nov 26, 1999 Consider a system where: Y1 = X0 + (1-g11)b1 X1 + g21b2X2 + g31b3X3 Y2 = X4 + g12b1 X1 + (1-g22)b2X2 + g32b3X3 Y3 = X5 + g13b1 X1 + g23b2X2 + (1-g33)b3X3 and: 1 = Y1 + Y2 + Y3, 0 ³ Yi ³ 1.0 "i 1 = X0 + X4 + X5, 0 ³ Xi ³ 1.0 "i 1 = g11 + g12 + g13, 0 ³ gij ³ 1.0 "i,j 1 = g21 + g22 + g33 1 = g31 + g32 + g33 GOAL: To estimate g's and b's. Problems: overidentification; effects of the data items and some of the parameters summing to 1 which I still don't fully understand. Background: This is for a voting transition study in Italy, where the b's represent a the proportion of voters following a rational proximity model, and the g's represent the discrete probability distribution according to which non-rational voters distribute their votes to one of three electoral coalitions (corresponding to the Y's). I have data for all of the Y's and X's, which are proportions. Possible ways to simplify: Set gij's to constants before estimation. Set gij = g* " i,j. Set b3 = 1. Set b1 = b2. File translated fromTEXby http://hutchinson.belmont.ma.us/tth/">TTH,version 2.56. On 26 Nov 1999, 14:00. --0FE2A80BBDB78FA4819CDE7F--
Re: bayes vs frequentist
You won't find the difference very striking when you look at what most statisticians do in practice. The differences between statistical practices of frequentists and Bayesians are rather technical. The difference lies in how they think about the results. Dick
Re: Some "elementary" issues -
The question of compatibility of two data sets is not answered by statistical tests. Two data sets are compatible if the "same procedure" was used for each. Any evidence of differences between the data sets is attributed either to differences between the material or differences between the actual procedures used in either case. No electronic database with automated procedures can answer such questions. A careful review of the observational procedures is where we start. Dick