Search found 3337 matches

by startz
Mon Aug 13, 2018 8:42 am
Forum: Estimation
Topic: logit standard error
Replies: 3
Views: 113

Re: logit standard error

I think you've done everything correctly. But you may be misinterpreting significance.

Suppose c(1) is almost exactly zero, so not "significant." Then

1/(1+exp(c(1))) = 1/(1+e^0)=1/(1+1) = 0.5

which is significantly different from zero.
by startz
Thu Aug 09, 2018 5:37 am
Forum: Estimation
Topic: Functional form in Box estimation
Replies: 2
Views: 50

Re: Functional form in Box estimation

Code: Select all

Y = (X^c(1))*(Z^c(2))
by startz
Fri Aug 03, 2018 12:22 am
Forum: Data Manipulation
Topic: Marginal effects multinomial logit model
Replies: 7
Views: 671

Re: Marginal effects multinomial logit model

E-11 is scientific notation for times 10 to the minus eleventh power. It is not an error.
by startz
Tue Jul 31, 2018 10:16 am
Forum: Estimation
Topic: logit standard error
Replies: 3
Views: 113

Re: logit standard error

It’s a touch messy. Basically var(f(b)) = f’(b)var(b)f’(b), where f’ means first derivative.

Probably the easiest way to get the standard error is to look under coefficient,views,wald test and test

Code: Select all

1/(1+exp(c(1)))=0
by startz
Fri Jul 27, 2018 1:26 am
Forum: Estimation
Topic: Problem in Automatical ARIMA Forecasting
Replies: 19
Views: 4873

Re: Problem in Automatical ARIMA Forecasting

Forecast them.
by startz
Tue Jul 24, 2018 8:04 pm
Forum: Programming
Topic: Program to run with existing workfile
Replies: 12
Views: 190

Re: Program to run with existing workfile

I am not shying from doing work, I am having a difficult time trying to find the instructions that tell the program to select an equation from an open workfile, change a given variable and then evaluate the equation for a given number of times and then store the SSR, similar to your example storing...
by startz
Tue Jul 24, 2018 3:12 pm
Forum: Programming
Topic: Program to run with existing workfile
Replies: 12
Views: 190

Re: Program to run with existing workfile

Presumably, something changed in the equation between runs. You'll have to reference the equation object after each run.
by startz
Tue Jul 24, 2018 2:40 pm
Forum: Programming
Topic: Program to run with existing workfile
Replies: 12
Views: 190

Re: Program to run with existing workfile

To expand on Gareth's answer...

Once you run an equation, it remembers all the values it produces until you change something. No need to re-run it. Also, you can run it once and save the value you want in a scalar for easy later reference.
by startz
Mon Jul 23, 2018 7:48 am
Forum: Programming
Topic: Multiplying Na by zero
Replies: 2
Views: 65

Re: Multiplying Na by zero

One way

Code: Select all

series newVar = @recode(oldvar1=na,0,oldvar1)*@recode(oldvar2=na,0,oldvar2)

or something like that, although that would also give NA*7 = 0 which may not be what you want.
by startz
Wed Jul 18, 2018 8:29 am
Forum: Estimation
Topic: Panel VAR EViews 9
Replies: 9
Views: 455

Re: Panel VAR EViews 9

Is it possible do do some of this by manually writing out the VAR as a standard regression?
by startz
Tue Jul 17, 2018 5:57 am
Forum: Estimation
Topic: Insufficient observations
Replies: 3
Views: 112

Re: Insufficient observations

EViews starts estimation with whatever is stored in the c vector. In your case, this is causing some division by zero. Try saying

Code: Select all

c = 1
before you start the estimation.
by startz
Mon Jul 16, 2018 9:10 am
Forum: Estimation
Topic: Near Singular Matrix Error
Replies: 39
Views: 30035

Re: Near Singular Matrix Error

Perhaps you should ask your instructor why normality matters in this application.

After all, lots of things just aren't normal.
by startz
Mon Jul 16, 2018 8:48 am
Forum: Estimation
Topic: Near Singular Matrix Error
Replies: 39
Views: 30035

Re: Near Singular Matrix Error

(1) You haven't said what the problem is. Non-normality does not invalidate a VAR.
(2) Sometimes you can take logs or make other transformations of the data to make residuals more normal.

Go to advanced search