Unit Root Test in EVIEWs

[@QRSchool]

If you have any query related please comment below

[@19meikinbekalimbekov33]

you said if prob is less than 5 % then it is non stationary but when interpreting you said that FDI is stationary at level and 1 difference but in 1 difference prob is not bigger than 5%.

[@doritos7134]

Dude you have helped me a lot for my econometrics ssignment, all your videos are relevant and go straight to the point!!!

[@suryadisamudra8097]

Very clear and straigh to the point, thanks..

[@hyeldabathonati3156]

you just made it simple and easy to understand. Thank you.

[@dipeshbanepali2557]

Excellent video

[@rizwanakhtar2296]

Very clear and good explanation, Thanks a Lot for the lecture

[@СохибЗиеев]

Thank you very much. I watched a lot of tutorials before your and found not understand result interests.

[@mohammedabbas8234]

Thanks mam,,, its really help me a lot. I am from Bangladesh.

[@johnmwakasege4581]

Excellent presentation, thanks...

[@yinyi3934]

Hi, may i know why you are using maximum lags of 7 in the unit root test?

[@godwinomokhua9165]

This very helpful, thank you

[@reggierosario4708]

Excellent explanation !! Thanks !

[@sakhawatemon4776]

How that table came from the graph. What are the calculations that gave the values? 8:42 .

[@QRSchool]

For tables, we used multiple methods like ADF PP Fisher etc, and graph is just the graphical representation of the rawdata.

[@saikrishna2940]

for logistic regression 1 and 0 we need to check this?

[@QRSchool]

If the dependent variable is dichotomous, logistic regression can be applied. For example: Gender is dependent variable, which has normally two options (Male & Female), it makes it dichotomous variable.

[@saikrishna8015]

@@QRSchool is any way to contact?

[@MuhammadAdli-px3br]

Hi, I want to ask. I have 4 variables. Dependent is index value and the other 3 independent variables is in money value (billion). May I know, should I change all of the variables value into change in percentage or log the 3 independent values and keep the index value? I plan to use eviews.

[@QRSchool]

If the unit of measures of variables are different (like in your case), simply take a log and do the analysis

[@muhammadadli3985]

@@QRSchool thank you, really appreciate your reply 👍🙏

[@izzatizulfakar]

@@QRSchool what if the variable has negative numbers? can we just take the absolute numbers and then log the variable, only then run a test?

[@cssunita3463]

Excellent

[@divyaganirashmika8283]

Very useful

[@usman37411]

What if the one value under level or 1st difference is less than 5 and another is more than 5 than how do we explain it? (For example IMP at level C=0.05 but C&T=0.50)

[@QRSchool]

This is called mix level of integration. At first difference, at both trend and trend & intercept, the trend should not be there.

[@j.a.cukwunna-mahori281]

Wonderful presentation!!! But pls kindly review your statement regarding the stationarity of the FDI data series with Prob = 0.072 which is greater than 0.05 (5%). Whereas Ho: Data series are non-stationary. Whereas also, decision rule states: reject Ho if Prob ≤ 5%(0.05) Pls, how then did you conclude that FDI is stationary?

[@larryanifowosed7628]

Hello , if a candidate is using primary data is it compulsory to do unit root test before going to use regression

[@QRSchool]

Only for secondary data, unit root analysis can be used. It is not used for primary as the data is cross-sectional.

[@takreembaig135]

i have a query. If unit root test shows the P value greater than 0.05 what should we do and what will be the impacts of that value.

[@QRSchool]

If you are using financial/economics data. There you can use significance level up to 10%.

[@madarajbyessien1423]

Many thanks for this video. I've learnt a lot

[@micheledeglinnocenti3384]

Excellent lecture you've done. A question please, i'm working about a financial project work that consists by analizying Ford Motors stock listed. How do i generate specific unit root test with private asset class's return ?

[@QRSchool]

The same way, it is displayed in the video. Generate a variable Asset Class return and check the trend.

[@madhvikaushik5297]

Nice wonderful explanation Thanku

[@QRSchool]

Thanks and welcome

[@usmansaleem1253]

In case Fri we can use both means level and 1st difference? Is there any issue for ardl or Johansen orany technique that some are stationery with intercept and some are with trend and intercept?

[@QRSchool]

For JJ cointegration, all variables should have same level of integration.

[@natureofbangla3731]

Mam can I convert variables into log form before doing unit roo t test and can I convert dependent variable into log form as well ??? Please

[@QRSchool]

Yes, if you intent to use log variables in analysis then initially transform the variable and use transformed variable into different analysis.

[@saikrishna8015]

In my study, all the variables are in level on trend and intercept. I am doing logistic regression. Is there any problem?

[@beautytipsbyhaleemasadia2547]

Great

[@emilinaayut9399]

If the time series is non stationary, what we should do with the time series before we do regression? Adjust the data or what?

[@QRSchool]

For regression, it is important that data series of variables should be stationary. We can change the measurement of the data. Even then, it is non-stationary then in this case regression is not appropriate.

[@yonasfekadu4169]

Thank you my dear for all your work but could you tag the excel file to practice it will be much helpful

[@QRSchool]

The data can be easily downloaded from WDI (World Bank Indicators). It's free and you can download data of any country. Even then you need an excel sheet, I can do that.

[@nurulraudhah1550]

what is the difference between intercept and (intercept & trend)?

[@QRSchool]

These are different options in Unit Root test. While testing stationarity status of the variables, we check unit root at intercept level, intercept first and so on....

[@sisyL17]

Thank you for the video, but when the data is stationary in first difference, does it mean that further analysis is carried out in the form of first difference?

[@QRSchool]

Yes, you will use first difference series in further analysis.

[@19meikinbekalimbekov33]

@@QRSchool So then it means we need to use only 1 difference data for further analysis? And how to do it?

[@aliefimei6509]

What means by I(0) and I (1)? How it can determine ARDL n VECM?

[@QRSchool]

I(0) represents level series and I(1) represents first difference series. For understanding ARDL and VECM, I will share another videos in coming days. Keep watching and Good Luck!

[@mstshilakhatun7850]

Thanks...It's really helpful

[@javeriajavaid3087]

what does it mean if beta is equal to 0 at all level, 1st difference and 2nd difference?

[@QRSchool]

Are you talking about Beta or Probability values? In Unit Root test, we do not have beta coefficients.

[@brishtichakraborty1193]

Mam, can I do unit root test after converting a variable into log variable? Answer me please

[@QRSchool]

Yes, you can!

[@Deepikabisht337]

unit root test is used for all the variables, i.e, IVs and DV , all.

[@QRSchool]

Yes. Trend should be checked from all the variables, regardless of their type.

[@vineshmolewa3696]

Aren't we suppose to log the variables before running the Unit Root Test on Eviews?

[@QRSchool]

Not necessary all the time. We use log series when we have to transform the series. Sometimes, the series of variables are measured in different units, so in this case log can be used.

[@izzatizulfakar]

@@QRSchool that means, if i have 4 variables, only 3 of them are in percentage, but one of them is in money value, then the variable that is in the money value needs to be changed to log before running the unit root test?

[@fahimahmed5439]

hello, I'm having some trouble. I created work file, but when I tried to run the unit root test this pop up appears: "Error unable to compute any results with the selected options". What should I do now? #QRSchool

[@QRSchool]

Kindly please elaborate, whether you have imported the data? which stage are you in?

[@saimanazirkhan7460]

Can you please mention link from where you collect data.

[@QRSchool]

I have used sample data

[@takreembaig135]

Sister would you please explain the transformation of variables into natural log?

[@QRSchool]

I will upload the video of the mentioned topic in the upcoming days. Please subscribe the channel so that you will be informed about all the new uploaded videos.

[@aliefimei6509]

What means by I(0) and I (1)? How it can determine ARDL n VECM?

[@saurabhkumarverma9263]

hello maam I need your help with research work

[@QRSchool]

Sure, you can ask specific question.

[@ishaqahmad5411]

Difference between trend and fluctuation???

[@QRSchool]

Trend means that the data is moving in the same direction whether increasing or decreasing. Whereas, fluctuation means that data is reverting to the mean, i.e. the central point.

[@karrinnr]

why in my eviews 10 there is no critical value test?

[@QRSchool]

You can take the decision with Prob value

[@nurnabilah8288]

Why didnt you do the unit root test with "none"?

[@QRSchool]

We can check the 'None' option as well. If you have a prior understanding of your data then choose any one which is appropriate, otherwise, choose all of three. But make sure all three results should be in-line.

[@dilipbjha]

@@QRSchool few questions 1) what should be the decision rule when there is conflicting result from various test (intercept, intercept +slope, none). In the given example, for first level data( FDI) intercept is greater than .05 but intercept & slope less that .05. 2) how to determine which method of unit root testing (int, int+slope, none) is appropriate for the given data set?

[@zoyashah7826]

Can u please tell me how to calculate the changed base year of GDP??

[@QRSchool]

Can you please rephrase the question. I think you are asking that how to calculate a difference series of GDP? Is it? Please confirm, so that I can answer accordingly.

[@zoyashah7826]

Yes how to recalculate the GDP using base yr as 2011..

[@aqsaali5688]

Plz interpret t values Jzak Allah

[@QRSchool]

With the help of ''t'' values, we can check significance level. If t value is greater than 1.95 (5% significance level), it means relationship is significant.

[@jp-xv7ip]

why did you make 0, ...... by FDI

[@QRSchool]

Please rephrase the question. I didn't get what you are asking?

[@mayurashetty3034]

Good explanation, can we download a crack version of Eviews in laptop? Does that cause any problems?

[@QRSchool]

Yes you can, but be careful from where to download because a lot can cause problems.

[@rizka_khr]

how can u decided to choose the maximum lags is 7? is there any requirement to choose it?

[@QRSchool]

I have not used 7 lags. Kindly specify, which step you are talking about?

[@irenet1751]

Why do you say that the FDI data series is stationary at level? c is 7,2%, that is more than 5%.

[@QRSchool]

When we are dealing with financial/economics kind data, there the significance level can be increased up to 10%. Lots of good research papers recommended this level.

[@irenet1751]

@@QRSchool Thank you. Are there any specific criteria we can use to decide that?

[@QRSchool]

@@irenet1751 We can take the reference from the previous relevant researches. Especially the researches that are published in renowned journals.

[@ArshadKhan-pm7td]

What the meaning of None

[@amnaarslan1751]

Mam plz tell me how to import data in unit root test from ms excel .plz tell me the link of the video where you have explained how to import in unit root test

[@QRSchool]

Watch kzbin.info/www/bejne/pmq8q6OLnMSLi6c

[@devarajan4016]

the null can rejected only when the prob is more to 0.05 or 5% right?.... how / why less than is rejected

[@QRSchool]

Yes, when probability is less than 5% (0.05), null hypothesis will rejected. It is only correct, when the confidence interval is 95%.

[@NaijaChinco]

Excellently clear. Thank you!