You may find it easier to use the logspline density fits (logspline
package) rather than the kernel density estimators for this.
On Mon, Oct 1, 2012 at 7:46 AM, Eugene Kanshin wrote:
> Hello,
> I have a data x with normal (or very close to normal) distribution, I can
> plot a density distribution
On Oct 1, 2012, at 6:46 AM, Eugene Kanshin wrote:
> Hello,
> I have a data x with normal (or very close to normal) distribution, I can
> plot a density distribution with density(x,...). My question is is there
> any way to calculate an area under this distribution (=probability) for
> particular
Forget density(). Smooth the ecdf instead.
?lowess
?smooth.spline
?monotone.smooth (in package fda, for monotone smoothing, which may be
preferable)
?locfit (in package locfit)
... and many many others
-- Bert
On Mon, Oct 1, 2012 at 6:46 AM, Eugene Kanshin wrote:
> Hello,
> I have a data x wit
Hello,
I have a data x with normal (or very close to normal) distribution, I can
plot a density distribution with density(x,...). My question is is there
any way to calculate an area under this distribution (=probability) for
particular range of x values, let's say for x from 0 to 2? I was not able
On Aug 24, 2009, at 8:11 AM, maram salem wrote:
Hi all,
I've a trivial question. If (q) is a continous variable,actually a
vector of 1000 values. how to calculate the probability that q is
greater than a specific value, i.e. P(q>45)??
sum(q>45)/1000 # if no NA's in vector
sum(q>45, na.rm
maram salem wrote:
Hi all,
I've a trivial question. If (q) is a continous variable,actually a vector of 1000
values. how to calculate the probability that q is greater than a specific value,
i.e. P(q>45)??
Do you want to estimate any distribution or do you just want the
empirical informati
Hi all,
I've a trivial question. If (q) is a continous variable,actually a vector of
1000 values. how to calculate the probability that q is greater than a specific
value, i.e. P(q>45)??
Thanks
Maram
[[alternative HTML version deleted]]
__
7 matches
Mail list logo
