On the robustness of LASSO

Probably the last post on lasso, before the summer break… More specifically, I was wondering about the interpretation of graphs \lambda\mapsto\widehat{\beta}_\lambda. We use them for variable selection, but my major concern was about confidence intervals : how can we trust those lines ?

As usual, a natural way is to use simulations on generated datasets. Consider for instance

Sigma = matrix(c(1,.8,.2,.8,1,.4,.2,.4,1),3,3)
n = 1000
library(mnormt)
X = rmnorm(n,rep(0,3),Sigma)
set.seed(123)
df = data.frame(X1=X[,1],X2=X[,2],X3=X[,3],X4=rnorm(n),
              X5=runif(n),
              X6=exp(X[,3]),
              X7=sample(c("A","B"),size=n,replace=TRUE,prob=c(.5,.5)),
              X8=sample(c("C","D"),size=n,replace=TRUE,prob=c(.5,.5)))
df$Y = 1+df$X1-df$X4+5*(df$X7=="A")+rnorm(n)

One can use other simulations of datasets, and store the output

vlambda = exp(seq(-8,1,length=201))
lasso = glmnet(x=X,y=df[,"Y"],family="gaussian",alpha=1,
             lambda=vlambda,standardize=TRUE)
VLASSO[[s]] = as.matrix(lasso$beta)

To visualize confidence bands, one can compute quantiles

Q05=Q95=Qm=matrix(NA,9,201)
for(i in 1:nrow(Q05)){
  for(j in 1:ncol(Q05)){
    v = unlist(lapply(VLASSO,function(x) x[i,j]))
    Q05[i,j] = quantile(v,.05)
    Q95[i,j] = quantile(v,.95)
    Qm[i,j]  = mean(v)
  }}

and get get the graph

plot(lasso,col=colrs,"lambda"ylim=c(min(Q05),max(Q95)))
colrs=c(brewer.pal(8,"Set1"))
polygon(c(log(lasso$lambda),rev(log(lasso$lambda))),
          c(Q05[2,],rev(Q95[2,])),col=colrs[1],border=NA)
polygon(c(log(lasso$lambda),rev(log(lasso$lambda))),
        c(Q05[5,],rev(Q95[5,])),col=colrs[2],border=NA)
polygon(c(log(lasso$lambda),rev(log(lasso$lambda))),
        c(Q05[8,],rev(Q95[8,])),col=colrs[3],border=NA)

An alternative (more realistic on real data) is to use bootstrapped version of the dataset

id = sample(1:nrow(X),size=nrow(X),replace=TRUE)
lasso = glmnet(x=X[id,],y=df[id,"Y"],family="gaussian",alpha=1,
               lambda=vlambda,standardize=TRUE)


So far, it looks it’s working very well. Now, what if we have a smaller dataset

n = 100

On simulated new samples, we get


while the bootstrap version is

There is more uncertainty, clearly, but the conclusion is not ambiguous here.

Now, what about real data. Consider the following

chicago = read.table("http://freakonometrics.free.fr/chicago.txt",header=TRUE,sep=";")
tail(chicago)
   Fire   X_1 X_2    X_3
42  4.8 0.152  19 13.323
43 10.4 0.408  25 12.960
44 15.6 0.578  28 11.260
45  7.0 0.114   3 10.080
46  7.1 0.492  23 11.428
47  4.9 0.466  27 13.731

with one variable of interest (the number of fires, per unhabitants) and 3 features. We can here use bootstrap to generate samples, and then fit a lasso regression. On the original dataset, the regression is

X = model.matrix(lm(Fire~.,data=chicago))
 id = sample(1:nrow(X),size=nrow(X),replace=TRUE)
 vlambda = exp(seq(-4,2,length=201))
 lasso = glmnet(x=X[id,],y=chicago[id,"Fire"],family="gaussian",alpha=1,
               lambda=vlambda,standardize=TRUE)

And if we just plot lines \lambda\mapsto\widehat{\beta}_\lambda we get

Now, consider bootstrap samples.

for(s in 1:100){
  id=sample(1:nrow(X),size=nrow(X),replace=TRUE)
  library(glmnet)
  vlambda=exp(seq(-4,2,length=201))
  lasso=glmnet(x=X[id,],y=chicago[id,"Fire"],family="gaussian",alpha=1,
               lambda=vlambda,standardize=TRUE)
  plot(lasso,col=colrs,"lambda",lwd=.2,add=TRUE)}

We get here

The interpretation here is much more difficult

What about the order ?

N=matrix(NA,100000,4)
for(s in 1:100000){
  id=sample(1:nrow(X),size=nrow(X),replace=TRUE)
  library(glmnet)
  vlambda=exp(seq(-4,2,length=201))
  lasso=glmnet(x=X[id,],y=chicago[id,"Fire"],
               family="gaussian",alpha=1,
               lambda=vlambda,standardize=TRUE)
  N[s,]=names(sort(apply(as.matrix(lasso$beta),
        1,function(x) sum(x!=0))))}

The ordering that was obtained on the original dataset was the same in 56% of the scenarios,

mean(apply(N,1,function(x) paste(x,collapse="")=="(Intercept)X_1X_2X_3"))
[1] 0.5693

We can look at all the cases,

L=as.character(c(123,132,213,231,312,321))
Li=paste("(Intercept)X_",substr(L,1,1),"X_",
         substr(L,2,2),"X_",substr(L,3,3),sep="")
g=function(y) mean(apply(N,1,function(x) paste(x,collapse="")==y))
vL=unlist(lapply(Li,g))
names(vL)=L
barplot(vL,las=2,horiz=TRUE)

Standardization in LASSO

The lasso regression is based on the idea of solving\widehat{\mathbf{\beta}}_{\lambda}=\text{argmin}\lbrace -\log\mathcal{L}(\mathbf{\beta}|\mathbf{x},\mathbf{y})+\lambda\|\mathbf{\beta}\|_{\ell_1}\rbracewhere\Vert\mathbf{a} \Vert_{\ell_1}=\sum_{i=1}^d |a_i|for any \mathbf{a}\in\mathbb{R}^d. In a recent post, we’ve seen computational aspects of the optimization problem. But I went quickly throught the story of the \ell_1-norm. Because it means, somehow, that the value of \beta_1 and \beta_2 should be comparable. Somehow, with two significant variables, with very different scales, we should expect orders (or relative magnitudes) of \widehat{\beta}_1 and \widehat{\beta}_2 to be very very different. So people say that it is therefore necessary to center and reduce (or standardize) the variables.

Consider the following (simulated) dataset

Sigma = matrix(c(1,.8,.2,.8,1,.4,.2,.4,1),3,3)
n = 1000
library(mnormt)
X = rmnorm(n,rep(0,3),Sigma)
set.seed(123)
df = data.frame(X1=X[,1],X2=X[,2],X3=X[,3],X4=rnorm(n),
X5=runif(n),X6=exp(X[,3]),
X7=sample(c("A","B"),size=n,replace=TRUE,prob=c(.5,.5)),
X8=sample(c("C","D"),size=n,replace=TRUE,prob=c(.5,.5)))
df$Y = 1+df$X1-df$X4+5*(df$X7=="A")+rnorm(n)
X = model.matrix(lm(Y~.,data=df))

Use the following colors for the graphs and the value of \lambda

library("RColorBrewer")
colrs = c(brewer.pal(8,"Set1"))[c(1,4,5,2,6,3,7,8)]
vlambda=exp(seq(-8,1,length=201))

The first regression we can run is a non-standardized one

library(glmnet)
lasso = glmnet(x=X,y=df[,"Y"],family="gaussian",alpha=1,lambda=vlambda,standardize=FALSE)

We can visualize the graphs of \lambda\mapsto\widehat{\beta}_\lambda

idx = which(apply(lasso$beta,1,function(x) sum(x==0))<200)
plot(lasso,col=colrs,'lambda',xlim=c(-5.5,2.3),lwd=2)
legend(1.2,.9,legend=paste('X',0:8,sep='')[idx],col=colrs,lty=1,lwd=2)

At least, observe that the most significant variables are the one that were used to generate the data.

Now, consider the case that we standardize the data

lasso = glmnet(x=X,y=df[,"Y"],family="gaussian",alpha=1,lambda=vlambda,standardize=TRUE)

The graphs of \lambda\mapsto\widehat{\beta}_\lambda

The graph is (strangely) very similar to the previous one. Except perhaps for the green curve. Maybe that categorical are not simular to continuous variables… Because somehow, standardisation of categorical variables might be not natural…

Why not consider some home-made function ? Let us transform (linearly) all variable in the X matrix (except the first one, which is the intercept)

Xc = X
for(j in 2:ncol(X)) Xc[,j]=(Xc[,j]-mean(Xc[,j]))/sd(Xc[,j])

Now, we can run our lasso regression on that one (with the intercept since all the variables are centered, but y)

lasso = glmnet(x=Xc,y=df$Y,family="gaussian",alpha=1,intercept=TRUE,lambda=vlambda)

The plot is now

plot(lasso,col=colrs,"lambda",xlim=c(-6.7,1.3),lwd=2)
idx = which(apply(lasso$beta,1,function(x) sum(x==0))<length(vlambda))
legend(.15,.45,legend=paste('X',0:8,sep='')[idx],col=colrs,lty=1,bty="n",lwd=2)

Actually, why not also center the y variable, and remove also the intercept

Yc = (df[,"Y"]-mean(df[,"Y"]))/sd(df[,"Y"])
lasso = glmnet(x=Xc,y=Yc,family="gaussian",alpha=1,intercept=FALSE,lambda=vlambda)

Hopefully, those graphs are very consistent (and if we use those for variable selection, they suggest to use variables that were actually used to generate the dataset). And having qualitative and quantitative variable is not a big deal. But still, I do not feel confortable with the differences…

Convex Regression Model

This morning during the lecture on nonlinear regression, I mentioned (very) briefly the case of convex regression. Since I forgot to mention the codes in R, I will publish them here. Assume that y_i=m(\mathbf{x}_i)+\varepsilon_i where m:\mathbb{R}^d\rightarrow \mathbb{R} is some convex function.

Then m is convex if and only if \forall\mathbf{x}_1,\mathbf{x}_2\in\mathbb{R}^d, \forall t\in[0,1], m(t\mathbf{x}_1+[1-t]\mathbf{x}_2) \leq tm(\mathbf{x}_1)+[1-t]m(\mathbf{x}_2)Hidreth (1954) proved that if m^\star=\underset{m \text{ convex}}{\text{argmin}}\left\lbrace\sum_{i=1}^n \big(y_i-m(\mathbf{x_i})\big)^2\right\rbracethen \mathbf{\theta}^\star=(m^\star(\mathbf{x_1}),\cdots,m^\star(\mathbf{x_n})) is unique.

Let \mathbf{y}=\mathbf{\theta}+\mathbf{\varepsilon}, then \mathbf{\theta}^\star=\underset{\mathbf{\theta}\in \mathcal{K}}{\text{argmin}}\left\lbrace\sum_{i=1}^n \big(y_i-\theta_i)\big)^2\right\rbracewhere\mathcal{K}=\{\mathbf{\theta}\in\mathbb{R}^n:\exists m\text{ convex },m(\mathbf{x}_i)=\theta_i\}. I.e. \mathbf{\theta}^\star is the projection of \mathbf{y} onto the (closed) convex cone \mathcal{K}. The projection theorem gives existence and unicity.

For convenience, in the application, we will consider the real-valued case, m:\mathbb{R}\rightarrow \mathbb{R}, i.e. y_i=m(x_i)+\varepsilon_i. Assume that observations are ordered x_1\leq x_2\leq\cdots \leq x_n. Here \mathcal{K}=\left\lbrace\mathbf{\theta}\in\mathbb{R}^n:\frac{\theta_2-\theta_1}{x_2-x_1}\leq \frac{\theta_3-\theta_2}{x_3-x_2}\leq \cdots \leq \frac{\theta_n-\theta_{n-1}}{x_n-x_{n-1}}\right\rbrace

Hence, quadratic program with n-2 linear constraints.

m^\star is a piecewise linear function (interpolation of consecutive pairs (x_i,\theta_i^\star)).

If m is differentiable, m is convex if m(\mathbf{x})+ \nabla m(\mathbf{x})^{\text{T}}\cdot[\mathbf{y}-\mathbf{x}] \leq m(\mathbf{y})

More generally, if m is convex, then there exists \xi_{\mathbf{x}}\in\mathbb{R}^n such that m(\mathbf{x})+ \xi_{\mathbf{x}}^{\text{ T}}\cdot[\mathbf{y}-\mathbf{x}] \leq m(\mathbf{y})
\xi_{\mathbf{x}} is a subgradient of m at {\mathbf{x}}. And then \partial m(\mathbf{x})=\big\lbrace m(\mathbf{x})+ \xi^{\text{ T}}\cdot[\mathbf{y}-\mathbf{x}] \leq m(\mathbf{y}),\forall \mathbf{y}\in\mathbb{R}^n\big\rbrace

Hence, \mathbf{\theta}^\star is solution of \text{argmin}\big\lbrace\|\mathbf{y}-\mathbf{\theta}\|^2\big\rbrace\text{subject to }\theta_i+\xi_i^{\text{ T}}[\mathbf{x}_j-\mathbf{x}_i]\leq\mathbf{\theta}_j,~\forall i,j and \xi_1,\cdots,\xi_n\in\mathbb{R}^n. Now, to do it for real, use cobs package for constrained (b)splines regression,

library(cobs)

To get a convex regression, use

plot(cars)
x = cars$speed
y = cars$dist
rc = conreg(x,y,convex=TRUE)
lines(rc, col = 2)


Here we can get the values of the knots

rc
 
Call:  conreg(x = x, y = y, convex = TRUE) 
Convex regression: From 19 separated x-values, using 5 inner knots,
     7,    8,    9,   20,   23.
RSS =  1356; R^2 = 0.8766;
 needed (5,0) iterations

and actually, if we use them in a linear-spline regression, we get the same output here

reg = lm(dist~bs(speed,degree=1,knots=c(4,7,8,9,,20,23,25)),data=cars)
u = seq(4,25,by=.1)
v = predict(reg,newdata=data.frame(speed=u))
lines(u,v,col="green")

Let us add vertical lines for the knots

abline(v=c(4,7,8,9,20,23,25),col="grey",lty=2)

Summer School, Big Data and Economics

This week I will be giving a lecture at the  2018 edition of the Summer School at the UB School of Economics, in Barcelona. It will be a four day crash course, starting on Tuesday (morning).

Lecture 1: Introduction : Why Big Data brings New Questions
Lecture 2: Simulation Based Techniques & Bootstrap
Lecture 3: Loss Functions : from OLS to Quantile Regression
Lecture 4: Nonlinearities and Discontinuities
Lecture 5: Cross-Validation and Out-of-Sample diagnosis
Lecture 6: Variable and model selection
Lecture 7: New Tools for Classification Problems
Lecture 8: New Tools for Time Series & Forecasting

Some slides are available on github, and probably more interesting, I will upload a R markdown with all the codes.