Lag selection. Model One. STATA

[@joshperry3162]

Play this on speed 1.5 and it sounds normal

[@sayedhossain23]

Thank you Perry, I would like to invite you to join Hossain Academy Facebook at below link and post your question there. Thank you once again, Sayed Hossain from Hossain Academy. facebook.com/groups/hossainacademy/

[@mengistunega7520]

Interesting! I like your voice.

[@Dr_Shiny]

Dear Sir, this stata command "versoc" mostly used for Time Series data optimal lag selection. whats about Panel data?

[@cengizhanguler4994]

did you get any information about that my friend?

[@ryanwilson9283]

but you didn't mention the Choice of variables. Stata manual says something different " To test for cointegration or fit cointegrating VECMs, we must specify how many lags to include. Building on the work of Tsay (1984) and Paulsen (1984), Nielsen (2001) has shown that the methods implemented in varsoc can be used to determine the lag order for a VAR model with I(1) variables. As can be seen from (9), the order of the corresponding VECM is always one less than the VAR. vec makes this adjustment automatically, so we will always refer to the order of the underlying VAR. " Thanks for the video

[@lucaborghi3064]

Dear Sayed, first of all thank you for updating these videos. They are very useful! I have a question; I have to do ADF test for some time series on STATA and I have to pick the right number of the lags! As you showed in this video, I calculate AIC and other criterions. My models have 5 independent variables. but I have a doubt: at 3:32 you insert all variables in "dependent variables". why? and if I insert only one variable, the results are different from those I obtained with all variables. how many dependent variables do i have to insert? which is the criteria? ADF test regards only one variable though! thank you in advance

[@akhliddinismailov3766]

Professor, is this approach applicable for ARDL model to use for panel data with quarterly data?

[@montassarkahia3136]

hello sayed i don't understand something in vecm model which is the coefficient. so if i got a short run causality after using the wald test and i suppose that i chose lag 2 and coefficients of independant variable in first difference are c(2) for lag 1 and c(3) for lag 2. so the question is about which coefficient (c(2) or c(3)) that i use in my interpretation.

[@pce369]

Why, when it asks for Dependent Variable, do you include the Dependent Variable and the three Independent variables?

[@akhliddinismailov3766]

What about lag selection for DF unit root test ? Does this way works for it as well?

[@sainaveenbali]

if i have dummy will i have to include it as variable as well for this test

[@stanleywu8300]

Sorry,I want to ask you a question.Could Stata demostrate ARDL bounds test?

[@liuran921]

is there minimum oberservation using VAR model? I have got 32 obsertation and 7 variables. is that valid for using VAR model to do analysis?

[@fabiangonzalo1662]

Buenos días como se hace selección de rezago en data panel?

[@philipvandelinde4960]

Dear Mister Hossain, Thank you for all your great videos! I have a question regarding my own study: I'm currently working on a study which incorporates an ARDL-VECM Model and the Pesaran bound testing approach. I found out that all my variables are I(0) or I(1), which is necessary. Now I want to test for cointegraiton between the variabels in stata. I do not understand exactly how to select the number of lags for each different variable in the ARDL-VECM model. Or do I use the same lag length for all variables after performing the test you demonstrate in this video? I have one dependent variable and four independent variables in the (long-run) model. Would be great if you could help me out. Thank you!

[@sayedhossain23]

Hello...Thank you for comment. Indeed I have done nothing yet with ARDL, so unable to comment but have plan in future.

[@200andrade]

Dear Dr. your videos are very helpful! but i realize you don't have any videos about GMM, 2SLS methods on stata... in case you have so can you please share it

[@nayef666]

Hello sir, are these variables in first difference or in levels?

[@sayedhossain23]

It is on data at level

[@xiaochenzhou7068]

Dear Mister Hossain, can I use this method to select lag in unit root test?

[@elanurturkuz854]

+Xiaochen Zhou did ever get an answer?

[@jetsekuijvenhoven3256]

One of you got an answer? Would like to know it.

[@arushigupta4065]

How to select significant lags for a static panel in STATA?

[@sayedhossain23]

Dear Gupta, Thank you. I would like to invite you to join Hossain Academy Facebook at below link and post your question there. If I know the answer I shall respond. Thank you once again, Sayed Hossain from Hossain Academy. facebook.com/groups/hossainacademy/

[@shubhalalbanik7695]

But, how to select lag for unit root test in STATA..?

[@kevinpony3620]

hi, have you ever got an answer? would like to know

[@synncmaster]

What about lag selection for Granger casuality test?

[@sayedhossain23]

Thank you. I would like to invite you to join Hossain Academy Facebook Group (Data Analysis) at below link and join our group discussion about modelling. Thank you once again, Sayed Hossain from Hossain Academy. facebook.com/groups/hossainacademy/

[@thakersDotCom]

and sir one more question is how many lags do we choose when we have monthly data ? you are doing a wonderful job sir (y)

[@sayedhossain23]

Regarding monthly data I have not done anything yet.

[@vinothraja83]

hi sir, thanks for your valuable videos shared, i am surfing your videos for my project on my masters, i am in need of using VAR model, in order to use VAR model, should i run co-integration rank of VECM. i already checked unit root test and found my data is not stationary except one variable. and while running co-Integration i choosed lag as 3 and 4 and results varies accordingly. in order to choose which lag should be captured, i viewed this video and run the pre estimation as told and applied 8 and found major stars in 8 the row. and while choosing 8 lags in johanson model for co-integration i got error stating " matrix not symetric" and max it allows only 5. i am not sure what should i do. can you help sir

[@sayedhossain23]

Vinoth Raja Thank you. I would like to invite you to join Hossain Academy Facebook for greater interaction about economics, finance and econometrics with me. Thank you Sayed Hossain from Hossain Academy. Please join below and post your question.facebook.com/groups/hossainacademy/

[@hijabrao3106]

Vinoth Raja

[@fuadhassan5750]

I am wondering, How do you expect different results by running the same command on the same data set. This video will confuse many people.

[@sayedhossain23]

Thank you Hassan. I would like to invite you to join Hossain Academy Facebook at below link and post your question there. Thank you once again, Sayed Hossain from Hossain Academy. facebook.com/groups/hossainacademy/

[@hagit_zivhagai9904]

thank you

[@sayedhossain23]

So far I know there is no such thing which is minimum number of observation but higher the observation, closer to population value

[@sayedhossain23]

I have not done video about it yet

[@sayedhossain23]

Yes in future

[@sayedhossain23]

So far I know ARDL can be handled using Microfit, the software I do not have to make video.

[@stanleywu8300]

thanks you : )

[@natro_09]

just speak without typing....we can hear you

[@sabihamarine4445]

too much talk in an easy issue. disgusting!

[@sayedhossain23]

Thank you. I would like to invite you to join Hossain Academy Facebook Group at below link and join our group discussion. Thank you. Sayed Hossain from Hossain Academy. facebook.com/groups/hossainacademy/

[@MuhammadAhmad-bz8gj]

Dear Sir, this stata command "versoc" mostly used for Time Series data optimal lag selection. whats about Panel data?