.Last.value

.Last.value
The .Last.value command gets the value returned by the most recent R command. .Last.value is most useful when coding on the fly.

Example. Data for a generalized linear model (GLM) is generated and a GLM model is fit to these data. In the first use the .Last.value command, I obtained the random numbers used to generate x and used identical to verify they were the same. This demonstration is important: .Last.value is actually the last value, not just the last command run again. Later in the code, I used .Last.value to grab the glm output without retyping or calling the previous line.
> set.seed(5)
> N    <- 50
> x    <- rnorm(N)
> temp <- .Last.value
> identical(x, temp)
[1] TRUE
> 
> lp <- -2 + 3*x
> p  <- exp(lp) / (1 + exp(lp))
> y  <- ifelse(runif(N) < p, 1, 0)
> 
> X  <- seq(-5, 5, 0.01)
> LP <- -2 + 3*X
> P  <- exp(LP) / (1 + exp(LP))
> 
> png("LastValue.png")
> plot(X, P, xlim=range(x), ylim=0:1,
+      type="l", ylab="Probability")
> points(x, y, col=2)
> dev.off()
null device 
          1 
> 
> glm(y ~ x, binomial)

Call:  glm(formula = y ~ x, family = binomial)

Coefficients:
(Intercept)            x  
     -2.625        3.418  

Degrees of Freedom: 49 Total (i.e. Null);  48 Residual
Null Deviance:	    61.09 
Residual Deviance: 21.92 	AIC: 25.92 
> model <- .Last.value
> model

Call:  glm(formula = y ~ x, family = binomial)

Coefficients:
(Intercept)            x  
     -2.625        3.418  

Degrees of Freedom: 49 Total (i.e. Null);  48 Residual
Null Deviance:	    61.09 
Residual Deviance: 21.92 	AIC: 25.92 
> 
> summary(model)

Call:
glm(formula = y ~ x, family = binomial)

Deviance Residuals: 
    Min       1Q   Median       3Q      Max  
-1.8604  -0.2764  -0.0934   0.2606   2.7035  

Coefficients:
            Estimate Std. Error z value Pr(>|z|)    
(Intercept)  -2.6251     0.9312  -2.819  0.00482 ** 
x             3.4185     0.9865   3.465  0.00053 ***
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1 

(Dispersion parameter for binomial family taken to be 1)

    Null deviance: 61.086  on 49  degrees of freedom
Residual deviance: 21.916  on 48  degrees of freedom
AIC: 25.916

Number of Fisher Scoring iterations: 7

GLM model and simulated data

Tip. The .Last.value command is really useful when needing to obtain the value from a computationally heavy command that was accidentally not saved.

Leave a Reply