Quantcast

LinearRegression Results

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

LinearRegression Results

Saleh Shahinfar-2

Dear Eibe, 


I am trying to fit a linear regression on my data alongside with other numeric prediction to be able to compare the results. 

The results of Linear regression is as below. 


Attributes, A, S, and D are nominal attributes. I am a bit confused what does this models mean? more precisely what does this means? 


-2.2421 * S=17,74,38,39,23,48,6,2,51,15,16,14,60,19,50,26,3 +
      1.5076 * S=74,38,39,23,48,6,2,51,15,16,14,60,19,50,26,3 +
     -0.9759 * S=39,23,48,6,2,51,15,16,14,60,19,50,26,3 +
      0.6143 * S=23,48,6,2,51,15,16,14,60,19,50,26,3 +

Why it repeats the same attributes with different regression coefficient? and what would be the actual model for prediction look like? I would like to understand more about how does this algorithm work?

Sincerely,
Saleh





=== Run information ===

Scheme:       weka.classifiers.functions.LinearRegression -S 0 -R 1.0E-8 -num-decimal-places 4
Relation:     Wool_Weather_Pasture_13122016-weka.filters.unsupervised.attribute.Remove-R1-2,186-196-weka.filters.unsupervised.attribute.NumericToNominal-R1,6-7,17
Instances:    7501
Attributes:   183
              [list of attributes omitted]
Test mode:    10-fold cross-validation

=== Classifier model (full training set) ===


Linear Regression Model

Y =

      4.8046 * A=5,6,4,7,3,2 +
     -0.7314 * A=6,4,7,3,2 +
      0.3915 * A=7,3,2 +
      2.356  * A=2 +
     -0.2799 * SEX +
      0.0713 * R +
     -2.2421 * S=17,74,38,39,23,48,6,2,51,15,16,14,60,19,50,26,3 +
      1.5076 * S=74,38,39,23,48,6,2,51,15,16,14,60,19,50,26,3 +
     -0.9759 * S=39,23,48,6,2,51,15,16,14,60,19,50,26,3 +
      0.6143 * S=23,48,6,2,51,15,16,14,60,19,50,26,3 +
      0.1254 * S=6,2,51,15,16,14,60,19,50,26,3 +
     -0.2165 * S=2,51,15,16,14,60,19,50,26,3 +
      0.3353 * S=51,15,16,14,60,19,50,26,3 +
     -0.2221 * S=15,16,14,60,19,50,26,3 +
      0.5918 * S=16,14,60,19,50,26,3 +
     -0.9435 * S=14,60,19,50,26,3 +
      0.7667 * S=60,19,50,26,3 +
      0.1643 * S=19,50,26,3 +
     -0.2135 * S=50,26,3 +
      0.4573 * S=26,3 +
     -0.6307 * S=3 +
      0.3961 * D=30,26,50,60 +
     -0.6256 * D=26,50,60 +
     -0.0154 * LD 

 


_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: LinearRegression Results

Eibe Frank-2
Administrator
LinearRegression uses the (supervised) NominalToBinary filter to turn nominal attributes into binary attributes that have values 0 or 1. This is done before the linear regression model is built.

For example, let’s take

  A=7,3,2

which is an attribute in your model output. This new attribute will have value 1 if the original nominal attribute A takes on values 7, 3, or 2, and it will have value 0 otherwise. The coefficient is for this new binary attribute.

Cheers,
Eibe

> On 9/01/2017, at 4:53 PM, Saleh Shahinfar <[hidden email]> wrote:
>
> Dear Eibe,
>
> I am trying to fit a linear regression on my data alongside with other numeric prediction to be able to compare the results.
> The results of Linear regression is as below.
>
> Attributes, A, S, and D are nominal attributes. I am a bit confused what does this models mean? more precisely what does this means?
>
> -2.2421 * S=17,74,38,39,23,48,6,2,51,15,16,14,60,19,50,26,3 +
>       1.5076 * S=74,38,39,23,48,6,2,51,15,16,14,60,19,50,26,3 +
>      -0.9759 * S=39,23,48,6,2,51,15,16,14,60,19,50,26,3 +
>       0.6143 * S=23,48,6,2,51,15,16,14,60,19,50,26,3 +
>
> Why it repeats the same attributes with different regression coefficient? and what would be the actual model for prediction look like? I would like to understand more about how does this algorithm work?
>
> Sincerely,
> Saleh
>
>
>
>
> === Run information ===
>
> Scheme:       weka.classifiers.functions.LinearRegression -S 0 -R 1.0E-8 -num-decimal-places 4
> Relation:     Wool_Weather_Pasture_13122016-weka.filters.unsupervised.attribute.Remove-R1-2,186-196-weka.filters.unsupervised.attribute.NumericToNominal-R1,6-7,17
> Instances:    7501
> Attributes:   183
>               [list of attributes omitted]
> Test mode:    10-fold cross-validation
>
> === Classifier model (full training set) ===
>
>
> Linear Regression Model
>
> Y =
>
>       4.8046 * A=5,6,4,7,3,2 +
>      -0.7314 * A=6,4,7,3,2 +
>       0.3915 * A=7,3,2 +
>       2.356  * A=2 +
>      -0.2799 * SEX +
>       0.0713 * R +
>      -2.2421 * S=17,74,38,39,23,48,6,2,51,15,16,14,60,19,50,26,3 +
>       1.5076 * S=74,38,39,23,48,6,2,51,15,16,14,60,19,50,26,3 +
>      -0.9759 * S=39,23,48,6,2,51,15,16,14,60,19,50,26,3 +
>       0.6143 * S=23,48,6,2,51,15,16,14,60,19,50,26,3 +
>       0.1254 * S=6,2,51,15,16,14,60,19,50,26,3 +
>      -0.2165 * S=2,51,15,16,14,60,19,50,26,3 +
>       0.3353 * S=51,15,16,14,60,19,50,26,3 +
>      -0.2221 * S=15,16,14,60,19,50,26,3 +
>       0.5918 * S=16,14,60,19,50,26,3 +
>      -0.9435 * S=14,60,19,50,26,3 +
>       0.7667 * S=60,19,50,26,3 +
>       0.1643 * S=19,50,26,3 +
>      -0.2135 * S=50,26,3 +
>       0.4573 * S=26,3 +
>      -0.6307 * S=3 +
>       0.3961 * D=30,26,50,60 +
>      -0.6256 * D=26,50,60 +
>      -0.0154 * LD
>  
> _______________________________________________
> Wekalist mailing list
> Send posts to: [hidden email]
> List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
> List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html

_______________________________________________
Wekalist mailing list
Send posts to: [hidden email]
List info and subscription status: https://list.waikato.ac.nz/mailman/listinfo/wekalist
List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html
Loading...