(EViews10):Cointegration, Series are I(0)

After performing a stationarity test, there are three (3) likely outcomes: the series may turn out to be I(0), I(1) or a combination of both. So, what do you do next? This hands-on tutorial shows you what to do in EViews10 when the series are I(0), that is, level-stationary series.
Here is the link to the ex21-1.wf1 dataset (EViews file) used for this tutorial (endeavour to have a Google account for easy accessibility): drive.google.c...
Follow up with soft-notes and updates from CrunchEconometrix:
Website: cruncheconometr...
Blog: cruncheconomet...
Forum: cruncheconometr...
Facebook: / cruncheconometrix
KZbin Custom URL: / cruncheconometrix
Stata Videos Playlist: • (Stata13):Estimate and...
EViews Videos Playlist: • (EViews10):Interpret V...

[@yaworomarickutowogbe175]

You are best. Only the langage IS my matter. I am francophone. Good job

[@CrunchEconometrix]

Thanks, Yawo!

[@spicita]

👏🏾👏🏾👏🏾👏🏾 THANK YOU

[@CrunchEconometrix]

You are so welcome, Ray!

[@rashmibhandari6275]

Hi, thanks for this insightful. What needs to be done when one series is I(1) and other series is I(0).

[@CrunchEconometrix]

Hi Rashmi, on what to do, kindly watch my video on "This is how to specify ARDL models". Afterwards, watch my videos on the Bounds Test followed by other ARDL/ECM videos.

[@mdatiqullahkhan8360]

Hello Mam, Thank you for your such detailed video. Can we get the excel data file to run and check own understanding if you kindly allow to do?

[@CrunchEconometrix]

Hi Atiqullah, thanks for the positive feedback. Please know that some of my data files are free while some are available upon purchase. Kindly check to confirm cruncheconomtrix.com.ng/shop

[@vishnushankar5574]

hi mam, How do I select my dependent and independent variable for doing OLS i.e., variables are I(0) at levels?

[@CrunchEconometrix]

Vishnu, this video and others on ARDL Modelling address your query. Kindly watch them all, thanks

[@domingosnhamussua3070]

dataset no longer available, please update

[@CrunchEconometrix]

Hi Domingos, I deactivated the link. Due to abuse and unethical conduct Stata dofiles and some datasets used in my videos are no longer free but available on my website upon payment. Here's the link cruncheconometrix.com/view/datashop.php

[@vishnushankar5574]

Hi mam, but the results shows serial autocorrelation which mean the model fitted is spurious regression. how to interpret and overcome that?

[@CrunchEconometrix]

Vishnu, you can modify the lags and re-estimate the model. Thanks

[@vishnushankar5574]

@@CrunchEconometrix Mam...but for fitting OLS for I(0) data, we got serial correlation in result. You told to increase the lag. But how do we fix lag while doing OLS analysis in eviews.There is no option in software. How do we avoid that problem of serial correlation?

[@CrunchEconometrix]

Vishnu, please follow the guidelines shown for ARDL-OLS video for ideas on increasing lag length.

[@babaywakhe]

I love your videos! I just want to find out what the possible cause for this model that you just estimated to be spurious, since you said that R-squared should be lower than the D-W stat.

[@CrunchEconometrix]

Yes, Zama I mentioned that in the clip that the variables are ASSUMED to be stationarity in level to emphasize the need for OLS estimation. But as you rightly observed R2>DW indicates spurious regression. Thanks.

[@jinhengboey7442]

Is there a must on the ln?

[@CrunchEconometrix]

Hi Jin, your query is unclear. Kindly re-state. Thanks.

[@nomaswazi_tshabalala]

Hi Prof, can I use eviews for panel data?

[@CrunchEconometrix]

Hi Nomaswazi, yes you can.

[@rozianabaharin9222]

hi.. i have requested for access to the dataset on last friday.. but still havent received any feedbak from your side.. please assist me. tq

[@CrunchEconometrix]

Hi Roziana, obviously you did not check your email. I responded on Sunday. Thanks.

[@djallelaimar781]

Professor, can you please help us with a video about logarithmic transformation of negative values using E-views, i have been looking for answers for so long, Thank you in advance Professor.

[@CrunchEconometrix]

Hi Djallel, transforming negative values will drop off. Loss of observations. My advice: use them naturally. That's what I do.

[@djallelaimar781]

@@CrunchEconometrix thank you so much Professor, i really appreciate your help

[@jackylin6281]

Hi professor,. What if my variables are fractionally integrated, does the integration still work？how can I process their relationship？

[@CrunchEconometrix]

Kindly explain fractionally integrated. Thanks

[@jackylin6281]

that is I(d), 0

[@CrunchEconometrix]

And you have such variable?

[@pedromrfernandes]

what If I have I(2) and I(1) data series? They can't be cointegrated, right? Should I do anything about it?

[@CrunchEconometrix]

Change the I(2) variable.

[@pedromrfernandes]

@@CrunchEconometrix But, for instance, my variables are real pc gdp and loans/gdp (or deflated loans pc), which is an I(2) process. So, if I change that variable to become an I(1) process, I will have to differentiate it, thus changing the initial idea of my study, as I will now be analysing loans' (yoy) change effect (relationship) on GDP, instead of the level of loans/gdp (or deflated loans pc) effect on GDP, right?

[@pedromrfernandes]

What I mean is that, due to being different order processes, there seems to be a mismatch between these two variables, as a result of not having cointegration, thus not having a long-term relationship, just a short-term one.

[@CrunchEconometrix]

@@pedromrfernandes Please keep your queries simple and short. I told you that you cannot use I(2) series.

[@CrunchEconometrix]

@@pedromrfernandes I prefer simple and straight-to-the-point queries. I have already responded to you on this.

[@zaykoylithinthong9790]

Hello professor if the data stationary in 2 different what should we do

[@CrunchEconometrix]

Hi Zaykoy, you can use the Toda-Yamamoto procedure though I've never had cause for it.

[@zaykoylithinthong9790]

@@CrunchEconometrix thanks for reply me professor, I have more questions I have GDP growth, interest rate, exchange rate and inflation rate. 1. Which variable should take log for it? If after unit root test , GDP growth, interest rate and inflation rate are 1 st difference but exchange rate is 2 nd difference. Which model should I use?

[@CrunchEconometrix]

(1) these are rate variables, don't log; (2) Toda-Yamamoto model.

[@CrunchEconometrix]

(1) these are rate variables, don't log; (2) Toda-Yamamoto model.

[@zaykoylithinthong9790]

@@CrunchEconometrix thanks professor 😊 I want to know what kind variable that we should take log?

[@adnansaqib9468]

there is no sequence in your videos. Give reference of past vidoes

[@CrunchEconometrix]

Adnan, I indicate the prerequisite videos to watch. You can also do a search query on my Channel for the video of interest. Thanks for your understanding.

[@spicita]

👏🏾👏🏾👏🏾👏🏾 THANK YOU

[@CrunchEconometrix]

Any time, Ray!