User Matthias Schmidtblaicher

24 votes

Using k-fold cross-validation for time-series model selection

answered Oct 30, 2016 at 12:00

15 votes

Does BIC try to find a true model?

answered Oct 20, 2016 at 16:15

14 votes

Accepted

State space representation of ARMA(p,q) from Hamilton

answered Feb 10, 2017 at 18:25

8 votes

Statistics without hypothesis testing

answered Jun 2, 2018 at 18:15

7 votes

Using Hodrick-Prescott filters for analyzing and forecasting business time series

answered Oct 21, 2016 at 16:06

7 votes

Accepted

Why does the Null Hypothesis have to be "equals to" and not "greater than or equal to"?

answered Oct 30, 2016 at 18:44

6 votes

Accepted

How to implement the Generalized Method of Moments for the upper limit of a uniform?

answered Nov 10, 2016 at 15:19

5 votes

Model averaging approach -- averaging coefficient estimates vs. model predictions?

answered Nov 6, 2016 at 15:36

5 votes

Accepted

Coefficients of Principal Components Regression in terms of original regressors

answered Oct 23, 2016 at 17:44

4 votes

If shrinkage is applied in a clever way, does it always work better for more efficient estimators?

answered Jun 22, 2017 at 10:40

3 votes

If multi-collinearity is high, would LASSO coefficients shrink to 0?

answered Oct 20, 2016 at 21:22

3 votes

Accepted

Compute the p-value of Diebold-Mariano Test

answered Jan 22, 2017 at 18:53

3 votes

Accepted

Orthogonality in PCA vectors

answered Nov 19, 2016 at 16:05

1 vote

What percentage of data should I separate for validation?

answered Jan 17, 2017 at 17:22

1 vote

Is it cheaper to buy fixed amounts or amounts with fixed price of a fluctuating product?

answered Feb 12, 2017 at 2:19

1 vote

Accepted

Model Confidence set

answered May 3, 2017 at 18:28

1 vote

Time Series AR1, coefficient accuracy different from linear regression

answered Jun 12, 2017 at 10:22

1 vote

Difference between average and expected value

answered Oct 23, 2016 at 14:29

1 vote

More than one unbiased estimator for a single unknown parameter?

answered Oct 18, 2016 at 16:51

1 vote

Accepted

Manually calculating PCA rotation using original data and PCA projections (scores)

answered Nov 2, 2016 at 13:40

0 votes

Accepted

R: n.restarts vs n.sim for function gosolnp() in Rsolnp package

answered Feb 16, 2017 at 1:48

Stack Exchange Network

21 Answers

Using k-fold cross-validation for time-series model selection

Does BIC try to find a true model?

State space representation of ARMA(p,q) from Hamilton

Statistics without hypothesis testing

Using Hodrick-Prescott filters for analyzing and forecasting business time series

Why does the Null Hypothesis have to be "equals to" and not "greater than or equal to"?

How to implement the Generalized Method of Moments for the upper limit of a uniform?

Model averaging approach -- averaging coefficient estimates vs. model predictions?

Coefficients of Principal Components Regression in terms of original regressors

If shrinkage is applied in a clever way, does it always work better for more efficient estimators?

If multi-collinearity is high, would LASSO coefficients shrink to 0?

Compute the p-value of Diebold-Mariano Test

Orthogonality in PCA vectors

What percentage of data should I separate for validation?

Is it cheaper to buy fixed amounts or amounts with fixed price of a fluctuating product?

Model Confidence set

Time Series AR1, coefficient accuracy different from linear regression

Difference between average and expected value

More than one unbiased estimator for a single unknown parameter?

Manually calculating PCA rotation using original data and PCA projections (scores)

R: n.restarts vs n.sim for function gosolnp() in Rsolnp package