Data Conversion to log. Model One. EVIEWS

[@sayedhossain23]

Normally I convert all variables into natural log so that heteroscedasticity problem may not arise after running regression model.

[@gerrymorgan5137]

Ahh, this is really helpful. Thank you so much. Greetings from Indonesia :)

[@sayedhossain23]

You are welcome

[@shaloms9640]

Hi, thank you for the usual educative video. This message came out when I tried to do so-"log of non positive number". How can I convert negative values into log?

[@vusimanikubhayi2003]

Any solution to this problem?

[@joe4g93]

hi mr sayed. can i know why we need to convert data into log form?

[@pujaneeparanayapa3461]

Sir, In a model where there are control variables. Can we convert the main independent variable to the log form without converting the control variables and the dependent variable .

[@godknowsncube4636]

Sir, will it be appropriate if l convert the data to logs before doing any test like unit root test etc ?

[@sayedhossain23]

Thank you for a good suggestion

[@suliman7890]

sir my data are not normal ,,so i cant take log bec there are some valu in negative so what i do with my data to convert it into normal

[@fatimasubzwari6970]

i try to convert variable into log but due to negative values i am unable to do ?? tell me the remedy to it please

[@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/

[@noorhafifahnoriah1997]

Sir, if my variable is interest rates which the unit in percentage (%), do i need to convert the interest rates data in to log?

[@sayedhossain23]

Noorhafifah Noriah No need

[@noorhafifahnoriah1997]

Sayed Hossain my variables are not in the same unit. some of the variables in percent (%) and some of them in currency units ( $). can i proceed or i need to convert them first into a same units such as in percent (%)?

[@sayedhossain23]

It can be different no problem

[@noorhafifahnoriah1997]

Sayed Hossain thank you so much sir.. your answer help me a lot :)

[@sayedhossain23]

You are welcome

[@renn9236]

if the data is in negative value, how i going to log it?

[@sayedhossain23]

Hello....Negative data can not convert to log. Thank you Sayed Hossain from Hossain Academy

[@renn9236]

Sayed Hossain so that mean the negative data cannot be use to run any test right? is it any solution can solve the negative data or i must change the variable (if my data have negative value) ?

[@sayedhossain23]

Only solution if you do not convert it into log. Just run the variable as it is. However, we have a Hossain Academy Facebook group where we are discussing about all these issue. You can post your question there and see. I am giving you the link below. Thank You,facebook.com/groups/hossainacademy/

[@renn9236]

Sayed Hossain Thanks You Sir =)

[@obaisi7272]

is that possible to key in LN in eviews?

[@sayedhossain23]

It is log?

[@obaisi7272]

yes.or we just key in (log)?

[@sayedhossain23]

logy=log(y)

[@obaisi7272]

it is.at 1st i thought LOG and LN are 2 very different commands.

[@sayedhossain23]

They are different. We use LN which is natural log.

[@gparragu]

Nice tutorial, thanks

[@shikhagoyal6866]

sir plz tell me whether i have to apply ADF unit root test on original series or newly created logged series??

[@sayedhossain23]

Dear Goyal, Thank you. I would like to invite you to join Hossain Academy Facebook at below link and post your question there. Actually I am in that group and may help you. Thank you once again, Sayed Hossain from Hossain Academy. facebook.com/groups/hossainacademy/

[@khcherifi9699]

i'm in the same problem ?? did u find an answer???

[@israqassem]

Thank you sir, its very useful, but when i converted the variable all the numbers became (NA), what should I do؟ whats the problem here?

[@sayedhossain23]

Dear Isra, 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 certainly respond. Thank you once again, Sayed Hossain from Hossain Academy. facebook.com/groups/hossainacademy/

[@kamal1977219772]

Hi Sayed Hossain im a PhD student I need your help how to use trans log equation to estimate seemingly unrelated regression, please show me how to estimate the parameters thank you.

[@sayedhossain23]

+Kamalalkatib Alshatrawi Not doe it yet. However, 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/

[@kamal1977219772]

thank you so much

[@GURMEET369]

Sir, Also explain me how to convert the negative variables in to log using Eviews

[@sayedhossain23]

GURMEET369 Dear GURMEET369, Thank you. Please join Hossain Academy Facebook below for greater interaction with me regarding data analysis. Thank you Sayed Hossain from Hossain Academy facebook.com/groups/hossainacademy/

[@DiabyAboubacarFassassy]

why we should convert data to log Sir

[@sayedhossain23]

To remove serial correlation from model. I would like to invite you to Hossain Academy facebook group below to discuss about models: facebook.com/groups/hossainacademy/

[@khcherifi9699]

if you have a multiplicative model time serie you should convert to log

[@Wironstone]

It is helpful, thank you. It would better if you repeat less your sentences.

[@sayedhossain23]

Sir, Yes you are right. I shall follow. Any comments about videos? Thank you Sayed Hossain from Hossain Academy

[@ashoksriram111]

Thank you sir, its very helpful. even after conversion of log, my data is not normally distributed . what to do sir?

[@sayedhossain23]

+Ashok Sriram Increase the sample size could be second option

[@joe4g93]

thank you sir :)

[@aliumuterler9625]

Thank you.

[@rutkettu]

Thanks!

[@ImamPutranto]

just talk less n remove useless explanation