EViews.com

Since your equation is nonlinear in parameters you need to write it out in equation form, as shown in the documentation.

The easiest way to indicate that a parameter needs to be estimated is to call it c(1) or c(2), etc. That's how EViews knows it is a parameter rather than a data series.

All it means is that there is not strong statistical evidence that the response is different from zero.

open a group with both series and graph them.

You can't have fewer instruments than coefficients.

After the estimation, proc/make regime results will create a series for you.

Aktar wrote:Dear Moderator,

I would like to estimate the coefficent theta in a logistic functional form. My model is

y = exp(theta x) / (1+exp(\theta x) + e

How can I specify it in the box estimation?

Thank you,

Something like

The site, which is an excellent resource, was last updated a week ago.

You may have converged to a local optima. Try the starting values from the full model.

The t-test is two-tailed. Not quite sure what a two-tailed F-test is.

Maybe the parameter c(72) is important.

Maybe. But state space models are generally nonlinear in that they don't have closed form solutions, so a search algorithm is necessary.

Sometimes nonlinear models are hard to estimate. Try different starting values.

It does not mean that OLS is preferred. It means GMM or IV has a problem.

It is certainly possible that adding an invalid instrument would switch a rejection to a non-rejection. The way to test for validity of a set of over-identifying instruments, as you know, is to use the J-statistic.