Тёмный

SPSS for newbies: Interpretation of the coefficients in regression when using log transformations 

Phil Chan
Подписаться 36 тыс.
Просмотров 19 тыс.
50% 1

Опубликовано:

 

18 сен 2024

Поделиться:

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист
Посмотреть позже
Комментарии : 11   
@vikvin74_DataAndStats
@vikvin74_DataAndStats 8 лет назад
Love watching your videos. Very simple very applied and very thorough
@Alebtong1
@Alebtong1 11 лет назад
Nice video but it may help to tell users that DV = dependent variable and IV= independent . It took me about 2 minutes to figure it out
@fahimjaved614
@fahimjaved614 7 лет назад
very informative video, Same rule would apply for System GMM coefficients?
@aznurmmu9300
@aznurmmu9300 5 лет назад
is there any difference if we use log 10 and log value? How to use log value in smartpls ? My Dv is log value and IVs not..IVs use latent scores they are 2nd order construct
@meiyaowu2434
@meiyaowu2434 7 лет назад
Great video~I noticed when you transformed data, you use ln() instead of log(). Is there any difference between these two ways?
@PhilChanstats
@PhilChanstats 7 лет назад
ln is the natural log,; log is to the base 10
@0548318196
@0548318196 11 лет назад
ty
@valkyriescope
@valkyriescope 6 лет назад
Hi, does the interpretation of R2 change? Thanks!
@PhilChanstats
@PhilChanstats 6 лет назад
Hi. Without re-watching my video, the definition of R2 applies to Y and set of Xs.
@simre5476
@simre5476 8 лет назад
What you can do if you transform your data in log and its still not normal? thanks!
@PhilChanstats
@PhilChanstats 8 лет назад
+simon renauld You may have a theoretical mathematical relationship between your Y and Xs, and this will determine what transformation you will do. For example, if your mathematical model is that there is an exponential relationship between Y and X in the form Y proportional to exp(X), then you would log both sides...if you are basing the choice soley on statistical properties and want the error term to be normal, then you can do: transform the Y but not the X, or do not transform Y and transform X, or transform both Y and X. The box-cox transformation is a power transformation that includes the log transform. The standard BC transform is for Y, where all the Y points are bigger than zero. Look it up, and good luck with your transforms - may the BC transform your life ;)
Далее
ВЛАД А4 СКАТИЛСЯ
09:34
Просмотров 532 тыс.
Bike Vs Tricycle Fast Challenge
00:43
Просмотров 16 млн
Multiple Regression - Interpretation (3of3)
11:36
Просмотров 185 тыс.
Data Transformation and Linear Regression in SPSS
6:06
Polynomial regression using SPSS (July 2020)
25:36
Просмотров 21 тыс.
Mean centering in regression in SPSS
9:25
Просмотров 60 тыс.
ВЛАД А4 СКАТИЛСЯ
09:34
Просмотров 532 тыс.