## Search found 3484 matches

Thu Aug 08, 2019 6:09 am
Forum: Estimation
Topic: forecasting
Replies: 1
Views: 90

### Re: forecasting

Number of lags here means to use many lags, For example 3 would mean using x(-1) x(-2) and x(-3)
Tue Aug 06, 2019 10:17 am
Forum: Bug Reports
Topic: switchreg small thing
Replies: 0
Views: 72

### switchreg small thing

If you estimate a least squares equation, the equation remembers the entry in the equation specification field when you hit the Estimate button. Switchreg doesn't remember the entry.
Tue Aug 06, 2019 7:18 am
Forum: Estimation
Topic: Markov switch state probabilities forecast
Replies: 18
Views: 429

### Re: Markov switch state probabilities forecast

You are asking people to do your work for you. The best way to get help is to show what you've already done and explain where you are stuck.
Mon Aug 05, 2019 12:52 pm
Forum: Estimation
Topic: Funny state switching
Replies: 0
Views: 42

### Funny state switching

Here's a snippet of a switching regression, simple switching on postwar GDP growth. Note that the standard error of the regression is larger than the standard deviation of the dependent variable. Is that odd?
Capture.PNG (3.96 KiB) Viewed 35 times
Mon Aug 05, 2019 5:58 am
Forum: Estimation
Topic: lag operator
Replies: 1
Views: 67

### Re: lag operator

lagged inflation is just inflation(-1), no d().
Mon Aug 05, 2019 5:56 am
Forum: Estimation
Topic: Regression help
Replies: 3
Views: 93

### Re: Regression help

I believe that Excel takes logs.
Mon Aug 05, 2019 5:54 am
Forum: Estimation
Topic: GMM estimation error: Singular Matrix
Replies: 5
Views: 158

### Re: GMM estimation error: Singular Matrix

Yes, the problem is you have more coefficients than variables.
Sun Aug 04, 2019 6:55 am
Forum: Estimation
Topic: GMM estimation error: Singular Matrix
Replies: 5
Views: 158

### Re: GMM estimation error: Singular Matrix

It looks like you are trying to estimate four coefficients with only three right=hand side variables.
Sun Aug 04, 2019 6:48 am
Forum: Estimation
Topic: Regression help
Replies: 3
Views: 93

### Re: Regression help

Remember that regression equations include error terms. The error terms here are not the same.
Fri Aug 02, 2019 4:26 pm
Forum: Econometric Discussions
Topic: data stationary
Replies: 5
Views: 184

### Re: data stationary

If data has a unit root, then it can be problematic in a regression. Data which is trend stationary is not generally a problem. So you shouldn't need to detrend at all.
Fri Aug 02, 2019 10:11 am
Forum: Econometric Discussions
Topic: data stationary
Replies: 5
Views: 184

### Re: data stationary

I'm not sure why you want to detrend the series, but you could try regressing on both a trend and a trend squared,.
Thu Aug 01, 2019 4:19 pm
Forum: Bug Reports
Topic: Something wonky with FRED
Replies: 1
Views: 119

### Something wonky with FRED

Capture2.PNG (11.91 KiB) Viewed 77 times

gets
Capture3.PNG (4.73 KiB) Viewed 77 times
Wed Jul 31, 2019 6:08 am
Forum: Estimation
Topic: Markov switch state probabilities forecast
Replies: 18
Views: 429

### Re: Markov switch state probabilities forecast

If the last in sample probability is last_prob and the transition probabilities are p11 and p01, then I think you want something like

prob(T+1) = last_prob*p11 + (1-last_prob)*p01
prob(T+2) = prob(T+1)*p11 + (1-prob(T+1))*p01

etc
Tue Jul 30, 2019 5:52 pm
Forum: Estimation
Topic: Markov switch state probabilities forecast
Replies: 18
Views: 429

### Re: Markov switch state probabilities forecast

Have you tried Proc Make Regime Probabilities?
Tue Jul 30, 2019 4:51 pm
Forum: Estimation
Topic: Markov switch state probabilities forecast
Replies: 18
Views: 429

### Re: Markov switch state probabilities forecast

sure, in fact I can extract the probabilities but how can I test the probabilities if in the "forecast" function I can only get the series fitted? how do I test insample and outofsample without the forecast function? Well, the extracted probabilities are in-sample forecasts. If you want t...