ols_step_forward_p(...) problem using the funtion ols_step_forward_p

[Molina-Valero]

---

Hello,

Please, could you help me with the next issue?:

I am getting the next error using the function ols_step_forward_p:

Error in if (pvals[minp] <= penter) { : the condition has length > 1 In addition: Warning messages: 1: In anova.lm(fullmodel) : ANOVA F-tests on an essentially perfect fit are unreliable 2: In anova.lm(full_model) : ANOVA F-tests on an essentially perfect fit are unreliable 3: In anova.lm(full_model) : ANOVA F-tests on an essentially perfect fit are unreliable 4: In anova.lm(fullmodel) : ANOVA F-tests on an essentially perfect fit are unreliable 5: In anova.lm(full_model) : ANOVA F-tests on an essentially perfect fit are unreliable 6: In anova.lm(fullmodel) : ANOVA F-tests on an essentially perfect fit are unreliable

The code I am runing is:

model<-lm(W.ha~., data=data)
model_f<-ols_step_forward_p(model, penter=0.1)

[aravindhebbali]

Hi @Molina-Valero,

I think we have fixed a similar issue some time back but am not sure.. Can you install the development version from GitHub and check if you still get the error?

# Install development version from GitHub
# install.packages("devtools")
devtools::install_github("rsquaredacademy/olsrr")

I will look into it if the error still persists after using the development version of the package.

[Molina-Valero]

Hi @aravindhebbali

Thank you for your help, but still have the same issue after installing the develpment version (0.6.0).

Sorry

[aravindhebbali]

No issues.. While I look into this, let me know if you can share the data. It will help in debugging the error.

[Molina-Valero]

Thank you! I am going to share the data later

[aravindhebbali]

Sure

[aravindhebbali]

Please install the develop branch of olsrr from GitHub for the fix:

# Install development version from GitHub
# install.packages("devtools")
devtools::install_github("rsquaredacademy/olsrr@develop")

# load library (development version)
library(olsrr)

# data
data <- read.csv('data.csv')

# model
model <- lm(W.ha ~ ., data = data)

# stepwise selection: use p_val instead of penter
ols_step_forward_p(model, p_val = 0.1)
 Stepwise Summary                            
----------------------------------------------------------------------
Step    Variable        AIC        SBC      SBIC      R2       Adj. R2 
----------------------------------------------------------------------
 0      Base Model    709.912    712.428      NA    0.00000    0.00000 
 1      p.b.mode.z    702.752    706.526      NA    0.29695    0.26766 
 2      var.r         698.076    703.108      NA    0.45616    0.40887 
 3      n.pts         694.380    700.671      NA    0.56314    0.50357 
 4      mode.rho      686.046    693.595      NA    0.70642    0.65050 
----------------------------------------------------------------------

Final Model Output 
------------------

                               Model Summary                                
---------------------------------------------------------------------------
R                           0.840       RMSE                    103103.928 
R-Squared                   0.706       MSE                13161472395.381 
Adj. R-Squared              0.651       Coef. Var                   17.008 
Pred R-Squared             -0.196       AIC                        686.046 
MAE                     82828.714       SBC                        693.595 
---------------------------------------------------------------------------
 RMSE: Root Mean Square Error 
 MSE: Mean Square Error 
 MAE: Mean Absolute Error 
 AIC: Akaike Information Criteria 
 SBC: Schwarz Bayesian Criteria 

                                      ANOVA                                       
---------------------------------------------------------------------------------
                        Sum of                                                   
                       Squares        DF         Mean Square      F         Sig. 
---------------------------------------------------------------------------------
Regression    665061288539.530         4    166265322134.883    12.633    0.0000 
Residual      276390920302.999        21     13161472395.381                     
Total         941452208842.529        25                                         
---------------------------------------------------------------------------------

                                              Parameter Estimates                                               
---------------------------------------------------------------------------------------------------------------
      model             Beta      Std. Error    Std. Beta      t        Sig             lower            upper 
---------------------------------------------------------------------------------------------------------------
(Intercept)    -18071950.603     5929827.602                 -3.048    0.006    -30403702.181     -5740199.025 
 p.b.mode.z        -9058.961        1636.669       -0.682    -5.535    0.000       -12462.601        -5655.321 
      var.r         6535.221        1995.734        0.390     3.275    0.004         2384.865        10685.577 
      n.pts            7.607           2.320        0.406     3.279    0.004            2.783           12.432 
   mode.rho    189752863.663    59272447.405        0.389     3.201    0.004     66489061.429    313016665.897 
---------------------------------------------------------------------------------------------------------------