Can you post the exact command you are typing and the exact
error that nfl2stata is returning?
↧
Re: Web scraping NFL data into Stata
↧
Re: Web scraping NFL data into Stata
nfl2stata scrape, season(_all)
Starting to pull data. This will take a while...
Processing 2009 game summary: [x/21 requests]
.........x.........x.
invalid 'Support'
↧
↧
Re: Export tables to Excel
I have a problem with tab2xl, it shows invalid syntax, and when I change the excel name, it begin says excel file open but i didn't open any excel file. Can someone help me?
↧
Re: Creating Excel tables with putexcel, part 2: Macro, picture, matrix, and formula expressions
any advice on how to put multiple outputs in one cell? ie a column of "Mean +/- SD"
↧
Re: Creating Excel tables with putexcel, part 2: Macro, picture, matrix, and formula expressions
or "Median (IQR)" etc
↧
↧
Re: Web scraping NFL data into Stata
I'm having this same issue for every year while it's scrapping.
↧
Re: Web scraping NFL data into Stata
". nfl2stata scrape, season(2018)
Starting to pull data. This will take a while...
Processing 2018 game summary: [x/21 requests]
.........x.........x.
invalid 'Support'
Processing 2018 profiles: [x/24 requests]"
When the processing 2018 profiles starts to happen it becomes really slow and nothing it showing even though it seems like something is loading.
Thank you
↧
Re: An introduction to the lasso in Stata
This is super clear. But one thing you do not explain is why you would prefer the post selection estimates to predict. Might you expand on when one should use the postselection estimates for this purpose, and when one should not?
↧
Re: An introduction to the lasso in Stata
Other question: are there rules of thumb one can use to determine how one should split the data into training and testing subsamples. Imagine one has a lot of data (say 500K observations). Crossvalidation takes a lot of time. If the training data was only 25%, things would go quicker. But what constitutes "enough" data to train the model?
↧
↧
Re: Introduction to Bayesian statistics, part 2: MCMC and the Metropolis–Hastings algorithm
This is a brilliant tutorial. I feel like I finally understood MCMC after days of trying. Thanks a lot!
↧
Re: An introduction to the lasso in Stata
Thank you for your question, Pierre. The post-selection coefficients are less biased than the penalized coefficients by definition. In the linear model, the post-selection coefficients may have better out-of-sample prediction performance (see Belloni, Alexandre, and Victor Chernozhukov (2013) )
However, there is no such theory for non-linear model.
↧
Re: An introduction to the lasso in Stata
As far as I know, there is no such rules of thumb to determine the ratio between training sample and testing sample. Intuitively, more training data should not hurt because the selected model may approximate the real world data well if my data are a good representative of the population.
↧
Re: Export tabulation results to Excel—Update
I keep getting the error message "option percentage not allowed" after I run the command:
tab2xl var1 sex [w=wt] using test, row(2) col(2) replace percentage
I updated to the latest command but I have not been able to get relative frequencies (%).
↧
↧
Re: Multilevel linear models in Stata, part 1: Components of variance
Hey,
I am writing on Master´s Thesis and wondering if it is also possbile to calculate the R² for three level models by just adding the variance coefficient of the third level in the formula.
I cannot find anything to this (random intercept model)
Thanks
↧
Re: Understanding omitted confounders, endogeneity, omitted variable bias, and related concepts
Hi, how i can gerate 4 variables in STATA that:
y is the dependent variable
x1 independet variable
D is a dummy(independent variable) that takes the value 1 if , for example have good job and 0 if bad job, but the D is in fact take the value 1 if omitted variable x2 is better educated and 0 if bad educated . Mening that
reg y x1 D , the D is biased due the education level(the true determinant)
note: the D should be generated as a dummy variable but X2 is sould be not a dummy variable.
would you help med please
↧
Re: Export tabulation results to Excel—Update
Hello Everyone.
I am currently using this package in stata 16. But it only works with oneway tabulations. I can't get it to work with twoway tabulations. It just says "invalid syntax" byt the syntax isn't invalid i am sure. And when i type percentage is says "option percentage not allowed".
Anyone who knows why this is?
↧
Re: Calculating power using Monte Carlo simulations, part 2: Running your simulation using power
Parts 3 (Linear & logistic regression) and 4 (Multilevel/longitudinal models) are now available at the links given below.
↧
↧
Re: Export tabulation results to Excel—Update
Run net install http://www.stata.com/users/..., replace. There was an issue with the current version for users.
↧
Re: Export tabulation results to Excel—Update
Run net install http://www.stata.com/users/..., replace. There was an issue with the current version for users.
FYI: The updated version does not support weighted percentages.
↧
Re: Export tabulation results to Excel—Update
This indicates that there is already an existing excel file handle. This usually happens if tab2xl gets into an error-state. To fix: run the putexcel close command.
↧