In: Statistics and Probability
SUMMARY OUTPUT | ||||||||
Regression Statistics | ||||||||
Multiple R | 0.40519 | |||||||
R Square | 0.164179 | |||||||
Adjusted R Square | 0.158531 | |||||||
Standard Error | 5.086625 | |||||||
Observations | 150 | |||||||
ANOVA | ||||||||
df | SS | MS | F | Significance F | ||||
Regression | 1 | 752.1843 | 752.1843 | 29.07132 | 2.69862E-07 | |||
Residual | 148 | 3829.316 | 25.87375 | |||||
Total | 149 | 4581.5 | ||||||
Coefficients | Standard Error | t Stat | P-value | Lower 95% | Upper 95% | Lower 95.0% | Upper 95.0% | |
Intercept | 37.21006 | 1.812045 | 20.53485 | 3.66E-45 | 33.6292408 | 40.79089 | 33.62924 | 40.79089 |
# Gears | -1.36117 | 0.252453 | -5.39178 | 2.7E-07 | -1.860051705 | -0.86229 | -1.86005 | -0.86229 |
1)
Regression model
highway fuel efficiency of a vehicle = 37.21006 - 1.36117 * Gears
2)
this model is statistical significant at alpha = 0.05 , becouse Significance F < alpha.
3)
Gears are statistical significant at alpha = 0.05 level and also intercept are significant at alpha =0.05.Both p-value are less than 0.05.
4)
The predicat highway fuel efficiency of a vehicle to increase by 37.21006 as per one unit of highway fuel efficiency of a vehicle.
The predicat highway fuel efficiency of a vehicle to Gears decrease by -1.36117 as per one unit of highway fuel efficiency of a vehicle.
5)
R-Square = 0.164179
16.42 % of the observed variation in highway fuel efficiency that
is explained by the model.It's mean's that the our model is not
good fit for given data .
6)
Given value Gears = 6
Regression model :
highway fuel efficiency of a vehicle = 37.21006 - 1.36117 * Gears
highway fuel efficiency of a vehicle = 37.21006 - 1.36117 * 6
highway fuel efficiency of a vehicle = 29.04304
..................................................................................................................................................................................
Note1; Our fit regression model is not good for given data please check the regression assumptions of regression equation. like, dependent variable are follow normal distribution, if not follow not normal distribution then used the log transformation.
a) outliear present it our data then remove it used the following formula.
mean+3*sigma
b) Remove the multicollinearity using the VIF = 1/(1- R-square ).
c) Also check the heteroscedasticity remove . used to if independent variable of each other are high correlation > 0.7 then remove it.
all step check it then fit Regression model for given data set. then improve our model accuracy.
..........................................................................................................................................................................
NOTE 2 : :If you want any additional information then comment please below i will provide ans.