predicaot2

interpretar.r

> m4 <- lm(geo$preco ~ geo$area + geo$coef_ap + geo$acl_decl + geo$frente, data=geo)
> 
> summary(m4)

Call:
lm(formula = geo$preco ~ geo$area + geo$coef_ap + geo$acl_decl + 
    geo$frente, data = geo)

Residuals:
   Min     1Q Median     3Q    Max 
-15525  -5235  -1858   3628  52939 

Coefficients:
               Estimate Std. Error t value Pr(>|t|)    
(Intercept)  -41771.930  16298.496  -2.563 0.013501 *  
geo$area         17.276      6.424   2.689 0.009761 ** 
geo$coef_ap    8195.227   2238.198   3.662 0.000614 ***
geo$acl_decl  55767.600  18004.224   3.097 0.003227 ** 
geo$frente      221.517     94.485   2.344 0.023152 *  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 11020 on 49 degrees of freedom
Multiple R-squared:  0.5867,	Adjusted R-squared:  0.5529 
F-statistic: 17.39 on 4 and 49 DF,  p-value: 6.111e-09

> summary(m4)

Call:
lm(formula = geo$preco ~ geo$area + geo$coef_ap + geo$acl_decl + 
    geo$frente, data = geo)

Residuals:
   Min     1Q Median     3Q    Max 
-15525  -5235  -1858   3628  52939 

Coefficients:
               Estimate Std. Error t value Pr(>|t|)    
(Intercept)  -41771.930  16298.496  -2.563 0.013501 *  
geo$area         17.276      6.424   2.689 0.009761 ** 
geo$coef_ap    8195.227   2238.198   3.662 0.000614 ***
geo$acl_decl  55767.600  18004.224   3.097 0.003227 ** 
geo$frente      221.517     94.485   2.344 0.023152 *  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 11020 on 49 degrees of freedom
Multiple R-squared:  0.5867,	Adjusted R-squared:  0.5529 
F-statistic: 17.39 on 4 and 49 DF,  p-value: 6.111e-09

> 
> summary(m4)

Call:
lm(formula = geo$preco ~ geo$area + geo$coef_ap + geo$acl_decl + 
    geo$frente, data = geo)

Residuals:
   Min     1Q Median     3Q    Max 
-15525  -5235  -1858   3628  52939 

Coefficients:
               Estimate Std. Error t value Pr(>|t|)    
(Intercept)  -41771.930  16298.496  -2.563 0.013501 *  
geo$area         17.276      6.424   2.689 0.009761 ** 
geo$coef_ap    8195.227   2238.198   3.662 0.000614 ***
geo$acl_decl  55767.600  18004.224   3.097 0.003227 ** 
geo$frente      221.517     94.485   2.344 0.023152 *  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 11020 on 49 degrees of freedom
Multiple R-squared:  0.5867,	Adjusted R-squared:  0.5529 
F-statistic: 17.39 on 4 and 49 DF,  p-value: 6.111e-09

> 
> residuos <- rstandard(m4)
> 
> preditos <- fitted(m4)
> 
> plot(residuos ~ preditos)
> 
> qqnorm(residuos)
> 
> qqline(residuos)
> 
> shapiro.test(residuos)

	Shapiro-Wilk normality test

data:  residuos
W = 0.82459, p-value = 1.653e-06

> 
> 

vars_transformadas.r

> coefap <- geo$coef_ap^2
> frente <- geo$frente^3
> acldecl <- 1/geo$acl_decl
> m4 <- lm(log(geo$preco) ~ log(geo$area) + coefap + acldecl + frente, data=geo)
> 
> summary(m4)

Call:
lm(formula = log(geo$preco) ~ log(geo$area) + coefap + acldecl + 
    frente, data = geo)

Residuals:
     Min       1Q   Median       3Q      Max 
-0.30260 -0.14413 -0.03326  0.09578  0.64391 

Coefficients:
               Estimate Std. Error t value Pr(>|t|)    
(Intercept)   10.546673   0.660423  15.970  < 2e-16 ***
log(geo$area)  0.227991   0.098045   2.325   0.0242 *  
coefap         0.041921   0.009218   4.547 3.58e-05 ***
acldecl       -1.534453   0.258429  -5.938 2.92e-07 ***
frente         0.003400   0.001822   1.867   0.0679 .  
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Residual standard error: 0.2031 on 49 degrees of freedom
Multiple R-squared:  0.709,	Adjusted R-squared:  0.6852 
F-statistic: 29.84 on 4 and 49 DF,  p-value: 1.35e-12

> 
> residuos <- rstandard(m4)
> 
> preditos <- fitted(m4)
> 
> plot(residuos ~ preditos)
> 
> qqnorm(residuos)
> 
> qqline(residuos)
> 
> shapiro.test(residuos)

	Shapiro-Wilk normality test

data:  residuos
W = 0.9426, p-value = 0.01198

> 
> 

MODELO 1 SEM TRANSFORMAR VARIÁVEIS

MODELO 2 TRANSFORMANDO AS VARIÁVEIS

Residuos vs Preditos (Sem Outliers)