EconBase
← Back to paper

Optimal selection of the number of control units in kNN algorithm to estimate average treatment effects

Extracted main text — title through conclusion, appendix excluded. This is what our citation measures are computed over, published so the extraction can be checked by eye.

41,654 characters · 5 sections · 29 citation commands

Rendered from LaTeX for readability, not typeset faithfully. Citation keys are highlighted; maths is left as source; figures, tables and equation environments are summarised rather than reproduced; unrecognised commands are greyed out so nothing is silently dropped. Email addresses are removed.

Optimal selection of the number of control units in kNN algorithm to estimate average treatment effects

\thispagestyle{empty}

abstract{10pt} We propose a simple approach to optimally select the number of control units in $k$ nearest neighbors (kNN) algorithm focusing in minimizing the mean squared error for the average treatment effects. Our approach is non-parametric where confidence intervals for the treatment effects were calculated using asymptotic results with bias correction. Simulation exercises show that our approach gets relative small mean squared errors, and a balance between confidence intervals length and type I error. We analyzed the average treatment effects on treated (ATET) of participation in 401(k) plans on accumulated net financial assets confirming significant effects on amount and positive probability of net asset. Our optimal $k$ selection produces significant narrower ATET confidence intervals compared with common practice of using $k=1$. { JEL Classification: C14, C18, O18, R42. \newline Keywords: Average treatment effects, k nearest neighbors, optimal number of control units, 401(k).}

\setcounter{page}{1} \pagenumbering{arabic}

Introduction

Matching estimators are commonly used to estimate average treatment effects when treatment assignment is independent of outcome conditional on confounding variables Cameron05. However, it seems that optimal selection of the number of matching units ($k$) is an open question as most of the applied literature fixes $k$ small without any optimal criterion, often $k=1$ Abadie2016. Choice of $k$ is a trade-off between bias and variance. Small $k$ implies small bias, but high variance, and vice versa. So, the aim of this paper is to propose a simple approach to optimally select $k$. In particular, we focus on the $k$ nearest-neighbors (kNN) algorithm when used to estimate treatment effects: average treatment effect (ATE), and average treatment effect on treated (ATET).

Optimal selection of $k$ in kNN algorithm is done by cross validation when prediction or classification are the objectives Hastie2009. This is easily performed as the statistical object of interest is observable in-sample. On the other hand, average treatment effects are not observable anywhere. So, we follow similar arguments to Athey2015 proposing an unbiased function of observable variables to optimally choose $k$ through cross validation when the main inferential concern is treatment effects.

We apply our proposal to analyze ATET due to enrollment in 401(k) on amount and probability of positive accumulated net financial assets. We found significant ATET whose point estimates are approximately \$15 K and 19% given an optimal $k^*=19$ in both cases.

After this brief introduction, Section (ref) shows our proposal to optimally select $k$ in the kNN algorithm. Section (ref) displays results from some simulations exercises, and Section (ref) shows the results of our application regarding 401(k) enrollment. Section (ref) ends with some concluding remarks.

Optimal selection of $k$ in kNN

We will focus on the average treatment effect on treated as we will use this in our application. However, results associated with the average treatment effects are in the Appendix (see subsection (ref)). We build the matching units using $k$ nearest neighbors (kNN) with replacement, where $k$ is selected minimizing the mean squared error between a conditional unbiased estimator of the average treatment effect on the treated (ATET) and a matching estimator of the ATET.

We consider a framework where there is a random binary treatment ($D_i=\left\{0,1\right\}$) that is independent of the outcome variable ($Y_i$) conditional on independent observable variables ($\bm{X}_i\subset \mathbb{R}^P$), $i=1,2,\dots,N$. The average treatment effect on treated,

equation[equation omitted — 102 chars of source]

where $\bm{X}_i$ are control variables, $\left\{Y_i(0),Y_i(1)\right\}$ are the potential outcomes under different states, untreated and treated,

align*[align* omitted — 112 chars of source]

However, we do not simultaneously observe the same individual under both states. So, we propose to use the matching approach to build the synthetic controls, the general formula is Cameron05,

equation[equation omitted — 157 chars of source]

where $A_k^i(\bm{x})=\left\{j:||\bm{X}_i-\bm{X}_j||<||\bm{X}_i-\bm{X}_l||, j\in\left\{D_j=0\right\}, l\in\left\{D_l=0\right\}\right\}$ is the set of the $k$ closest untreated units to treated unit $i$ in terms of the Euclidean norm in covariates, $j,l=1,2,\dots,N_0$, $||\cdot ||$ is the Euclidean norm, $N_0$ is the sample size of untreated individuals ($D_i=0$), $N_1$ is the sample size of treated individuals ($D_i=1$), and $k$ is the cardinality of $A_k^i(\bm{x})$, that is, the number of controls for treated unit,

equation*[equation* omitted — 134 chars of source]

Therefore, the synthetic control for treated individual $i$ is built using an average of untreated nearest individuals (neighbors).

The choice of $k$ is a trade-off between bias and variance. Selecting just one neighbor minimizes bias, this is equal to zero if there is exact match ($k=1, \bm{x_i}=\bm{x_j}$), but implies high variability. On the other hand, a large amount of neighbors decreases variance but increases bias. In general, $\bm{x_i}=\bm{x_j}$ is an event of probability 0 for continuous covariates, then Abadie2006 show that matching estimators have an asymptotic bias. This asymptotic bias can be ignored if $N_1^{P_c/2}=O(N_0)$ where $P_c$ is the number of continuous covariates ($P_c/2>1$). The bias of the ATET is

equation[equation omitted — 163 chars of source]

where $\mu_0(\bm{x}_i)=\mathbb{E}\left[Y|\bm{X}=\bm{x},D=0\right]$. This can be estimated non-parametrically using a series expansion estimator to obtain $\hat{B}^{TM}$ Abadie2011.

Therefore by equations (ref) and (ref), the bias-corrected matching estimator is

equation[equation omitted — 86 chars of source]

Abadie2006 show under suitable assumptions (see Assumptions 1, 2', 3' and 4 in their paper) that

equation*[equation* omitted — 127 chars of source]

where $V^{E,T}=\frac{1}{N_1}\sum_{i=1}^N \left(D_i+(1-D_i)\frac{J_k^i}{k}\right)^2 \sigma^2(\bm{X}_i,D_i)$, $J_k^i=\sum_{j=1}^N\mathbbm{1}\left\{i\in A_k^j\right\}$ is the number of times unit $i$ is used as a match given $k$ matches per unit, $\sigma^2(\bm{X}_i,D_i)=\mathbb{V}(Y|\bm{X}_i=\bm{x}_i,D_i=d_i)$, which can be consistently estimated by $\hat{\sigma}^2(\bm{X}_i,D_i)=\frac{k}{k+1}\left(Y_i-\frac{1}{k}\sum_{l\in L_k^i}\right)$, $L_k^i(\bm{x})$ is the set of the $k$ closest units to unit $i$ in terms of the Euclidean norm in covariates such that have the same value for the treatment variable, and $V^{\Delta(\bm{X}),T}=\mathbb{E}\left[(\left[\mu_1(\bm{X}_i)-\mu_0(\bm{X}_i)\right]-{\Delta}^{T})^2|\bm{D}=1\right]$ is the variance of the conditional average treatment effect on treated.

In an ideal situation, we select $k$ in the regression nearest neighbor framework (kNN) minimizing the mean squared error of the average treatment effect on treated (equation (ref)), that is,

equation[equation omitted — 116 chars of source]

However, the average treatment effect on treated ($\Delta^{T}$) is not observed. Therefore, we follow similar arguments to Athey2015 proposing an unbiased function of potentially observable variables to define the ATET in program (ref).

We have the following assumptions:

assumptionStable unit treatment value assumption (SUTVA)
assumptionIgnorability (unconfoundedness) \begin{equation*} D_i\perp Y_i(0) | \bm{X}_i=\bm{x}_i. \end{equation*}
assumptionOverlap (matching) \begin{equation*} 0\leq P(D_i=1 | \bm{X}_i=\bm{x}_i)<1. \end{equation*}
remarkAssumption (ref) implies not spillover, interaction or general equilibrium effects Rubin1978. Assumption (ref) establishes that treatment assignment ignores untreated outcome given control variables Rubin1978,Rosenbaum1983. Assumption (ref) says that it is necessary to have overlap in subsamples of $\bm{X}_i$, that is, for each treated individual it is necessary to have an analogous untreated (synthetic control) to identify the treatment effect on treated Rosenbaum1983.

We assume conditional independent and identically distributed random sample drawn from a population. This implicitly implies Assumption (ref) (SUTVA). Then, we define the individual treatment effect on treated

equation[equation omitted — 109 chars of source]

where $P\left[D=1\right]$ is the marginal probability of treatment, which is the same for every individual given random sampling, and $P(\bm{X}_i)=P\left[D_i=1|\bm{X}_i=\bm{x}\right]$ is the conditional probability of treatment given regressors $\bm{X}_i$, that is, the propensity score Rosenbaum1983.

propositionAssuming a conditional independent and identically distributed random sample such that Assumptions (ref) and (ref) are satisfied, then: \begin{equation*} \mathbb{E}\left[{\Delta_i^{T*}}\big| P(\bm{X}_i),\bm{X}_i=\bm{x}\right]=\Delta^{T}. \end{equation*}
proofWe omit conditioning on $P\left[D=1\right]$ and $\bm{X}_i=\bm{x}$ to simplify notation. \begin{align*} \mathbb{E}\left[{\Delta_i^{T*}}\big| P\left[D=1\right],\bm{X}_i=\bm{x}\right]=&\mathbb{E}\left[\frac{Y_i(D_i-P(\bm{X}_i))}{P\left[D=1\right](1-P(\bm{X}_i))}\right]\\ =&\frac{1}{P\left[D=1\right] (1-P(\bm{X}_i))}\left\{(1-P(\bm{X}_i))\mathbb{E}\left[D_i(Y_i(1)-Y_i(0))\right] \right.\\ & \left.+ \mathbb{E}\left[Y_i(0)(D_i-P(\bm{X}_i))\right]\right\}\\ =&\frac{1}{P\left[D=1\right] (1-P(\bm{X}_i))}\left\{(1-P(\bm{X}_i))\mathbb{E}\left[\mathbb{E}\left[D_i(Y_i(1)-Y_i(0))\big|D_i\right]\right] \right.\\ & \left. + \mathbb{E}\left[D_i\right]\mathbb{E}\left[Y_i(0)\right]-P(\bm{X}_i)\mathbb{E}\left[Y_i(0)\right]\right\}\\ =&\frac{1}{P\left[D=1\right] (1-P(\bm{X}_i))}\left\{(1-P(\bm{X}_i))P\left[D=1\right]\mathbb{E}\left[(Y_i(1)-Y_i(0))\big|D_i=1\right] \right.\\ & \left. + \mathbb{E}\left[Y_i(0)\right](\mathbb{E}\left[D_i\right]-P(\bm{X}_i))\right\}\\ =&\mathbb{E}\left[(Y_i(1)-Y_i(0))\big|D_i=1\right] \end{align*} The first equality is by definition, the second is after some simply algebra and taking into account that $Y_i=D_i Y_i(1)+(1-D_i) Y_i(0)$ and $D_i Y_i=D_i Y_i(1) + D_i Y_i(0) - D_i Y_i(0)$ given $D_i=\left\{0,1\right\}$. The third is an application of the law of iterative expectations for the first term, and the Assumption (ref) (ignorability) for the second term. The fourth equality takes again into account that $D_i=\left\{0,1\right\}$, and $P\left[D_i=1\right]=P\left[D=1\right]$ (random sampling). Finally, we take into account that $\mathbb{E} \left[D_i|\bm{X}_i=\bm{x}_i\right]=P\left[D_i=1|\bm{X}_i=\bm{x}\right]=P(\bm{X}_i)$. Notice that Assumption (ref) is required to have $P\left[D=1\right](1-P(\bm{X}_i))\neq 0$. \begin{remark} Observe that $\mathbb{E}\left[Y_i(0)D_i\right]=P\left[D=1\right]\mathbb{E}\left[Y_i(0)|D_i=1\right]$. Therefore, we also obtain the required statement, if \begin{equation*} \mathbb{E}\left[Y_i(0)|D_i=1\right]=\mathbb{E}\left[Y_i(0)|D_i=0\right]=\mathbb{E}\left[Y_i(0)\right] \end{equation*} conditional on $\bm{X}_i=\bm{x}_i$ is satisfied, that is, the conditional mean assumption on control group is satisfied Angrist1999. Outcome of untreated individuals does not determine participation. \end{remark}

We require estimators for $P\left[D=1\right]$ and $P(\bm{X}_i)$ to estimate $\Delta_i^{T*}$. We use $\frac{1}{N}\sum_{i=1}^N D_i$, $N=N_0+N_1$, as estimator for the marginal probability of treatment, and a logit model to estimate the conditional probability of treatment. Obviously, these estimators can be changed. Notice that this introduces an extra source of variability as we do not observe $P\left[D=1\right]$ and $P(\bm{X}_i)$ in observational data sets. This kind of issue is discussed by Athey2015 who point out that transformations like equation (ref) does not optimally use all available information, this may mean extra variability. Additionally, Abadie2016 show that ATE using matching based on the estimated propensity score is more efficient than matching based on the true propensity score. On the other hand, ATET using the estimated propensity score can be more o less efficient compared to the true propensity score. We performed some simulation exercises that show better inferential performance based on the estimated propensity score (available upon authors request). In addition, as our ATET (ATE) is asymptotically biased corrected, the mean squared error decomposition shows that variance explains the most part (75% on average), as expected. However, it remains a proportion to finite sample bias effect (25% on average).

Give that ${\Delta_i^{T*}}$ in equation (ref) is a conditional unbiased estimator for $\Delta^T$ (Proposition (ref)), we select $k$ in the kNN regression framework such that

equation[equation omitted — 194 chars of source]

where

equation*[equation* omitted — 230 chars of source]
equation*[equation* omitted — 195 chars of source]
equation*[equation* omitted — 218 chars of source]

and

equation*[equation* omitted — 74 chars of source]

where $G$ is the number of groups, $N_g$ is the sample size of each group, and $N_{1,g}$ is the sample size of the treated group in group $g$, $g=1,2,\dots,G$. Superscripts $Train$ and $Test$ refer to train and test data sets (see below).

Notice that $\widehat{\Delta_{i,g}^{T*}}$ is the sample version of the conditional unbiased estimator of the ATET $\Delta_i^{T*}$, so its average is a consistent estimator for ATET. On the other hand, $\hat{\Delta}_g^{BC}$ is bias-corrected version of the matching estimator Abadie2006.

kNN regression is a standard approach in the machine and statistical learning communities Hastie2009. So, we follow $k$-fold cross-validation to select the optimal $k$ Hastie2009. In particular, we randomly split the data set in $G$ roughly equal-sized groups such that we keep the proportion between treated and untreated individuals in each group, then we calculate $\frac{1}{N_g}\sum_{i=1}^{N_g}\widehat{\Delta_{i,g}^{T*}}$, $g=1,2,\dots,G$. We fit the model using $G-1$ groups, that is, we fix $k$, and identify the $k$ nearest neighbors in the control group at the training data set in terms of the Euclidean norm to individual $i$ in the treated group at the $G$-th left group (test data set), that is,

equation*[equation* omitted — 186 chars of source]

then we estimate $\frac{1}{k}\sum_{j\in A_k^i(\bm{x}_g^{Test})}Y_{j,g}^{Train}(0)$, and obtain $\hat{\Delta}_{i,g}^{TM}=Y_{i,g}^{Test}(1)-\frac{1}{k}\sum_{j\in A_k^i(\bm{x}_g^{Test})}Y_{j,g}^{Train}(0)$ and $\hat{\Delta}_{i,g}^{BC}=\hat{\mu}_0(\bm{x}_i)_{g}^{Test}-\frac{1}{k}\sum_{j\in A_k^i(\bm{x}_g^{Test})}\hat{\mu}_0(\bm{x}_j)_{g}^{Test}$ for $i\in\left\{D_{i,g}^{Test}=1\right\}$. So, we can calculate the mean squared error for the $G$-th left group. This is done for $g=1,2,\dots,G$ and different values of $k$ such that we average over the $G$ prediction errors given $k$, and select the $k$ which has the minimum average error (see program (ref)).

Given an optimal $k=k^*$, we use equation (ref) to estimate the average treatment effect on treated using the whole sample size. As we said, this estimator converges in distribution to a normal distribution, so we use the asymptotic result to build confidence intervals.

Algorithm (ref) summarizes our methodological proposal:

algorithm[algorithm omitted — 3,032 chars of source]

A Monte Carlos study

We set exactly the same simulation setting as Otsu2017 who proposed a weighted bootstrap to perform inference of matching estimators for ATE and ATET as it is well known that naive bootstrap produces invalid inference due to failing to reproduce the distribution of the number of times each unit is used as a match Abadie2008.

We show the simulation setting for exposition purposes:

align*[align* omitted — 467 chars of source]

where $\epsilon_i$, $v_i$, $\psi_i$ and $\zeta_i$ are mutually independent. The ATET as well as ATE is $\tau = 0.5$, $\gamma_1=0.15$, $\gamma=0.7$, $P=5$ and there are six different curves for $m_j(||\bm X_i||)$ (see Otsu2017 for details):

align*[align* omitted — 542 chars of source]

In addition, we perform other simulation exercises where the outcome variable is binary. In particular,

align*[align* omitted — 305 chars of source]

where $\epsilon_i\sim LG(0,1)$, $LG$ denotes a logistic distribution, $\beta=0.5$, $\tau= 0.5$, and $P(D_i=1)=P(\gamma_1+\gamma_2 ||\bm X_i||>v_i)=F_{LG}(\gamma_1+\gamma_2 ||\bm X_i||)$, $\gamma_1=0.15$, $\gamma_2=0.4$, $v_i\sim LG(0, 1)$ and $F_{LG}$ is the distribution function. Other components of the specification are as in the previous setting.

Observe that in this setting $ATE_i=\mathbbm{E}(Y_i(1)-Y_i(0)|\bm X_i=\bm x_i, Z_i = z_i)=P(Y_i(1)=1|\bm X_i=\bm x_i, Z_i = z_i)-P(Y_i(0)=0|\bm X_i=\bm x_i, Z_i = z_i)=F_{LG}(\tau + \beta m_j(||\bm X_i||))-F_{LG}(\beta m_j(||\bm X_i||))$, which has the same analytical expression for the ATET, but conditional on the relevant sample ($D_i=1$).

Our results are based on a sample size equal to 100 Otsu2017.\footnote{We perform other simulation exercises using larger sample sizes. Outcomes show similar results. Available upon authors request.} We have 1,000 replications for each simulation exercise.

First, we present the results for the average treatment effects on treated for the continuous outcome variable.\footnote{Results for the average treatment effects are in the Appendix, subsection (ref). Outcomes are similar to ATET.} Table (ref) displays the mean relative squared errors (MRSE),

equation*[equation* omitted — 168 chars of source]

where $\hat{\Delta}^{BC}_s(k)$ and $\hat{\Delta}^{BC}_s(k^*)$ are the ATET estimates using any $k$ and the optimal $k=k^*$, and $S_k$ is the number of times that $k$ was not optimal.

We see in Table (ref) that there are relative large advantages using our optimal $k^*$ as all figures are larger than 1.

table[table omitted — 2,328 chars of source]

Figure (ref) shows the average over 1,000 replications of the ratios between the length of the 95% confidence intervals associated with any number of neighbors ($k$, $x$ axis) and the optimal selection using our proposal. This is done for the six data generating settings ($m_j(||\bm X_i||), j=1,2,\dots 6$). We see in this figure that a small $k$ implies relative wide intervals, whereas large $k$ implies narrow intervals. This is expected as small $k$ implies high variability compared with large $k$. Observe that intervals lengths decreases very fast, and stabilize around $k=10$. On the other hand, we see in Figure (ref) that the type I error due to rejecting the null hypothesis $ATET=\tau$, given a significance level equal to 5%, is the lowest using $k=1$. Using $k$ minimizing the mean squared error gives a relatively small type I error compared with other $k$ selection.

Figure (ref) displays the histograms of the optimal $k$ in our six simulation settings. We see that optimal selection is dominated by small $k$. This could be evidence to support $k=1$, which is widely used in many applications. Although, it seems that $k=\left\{2,3\right\}$ are the most relevant, and large values of $k$ cannot be discarded, especially when using large sample sizes (we performed other simulations exercises that show this).

figure[figure omitted — 412 chars of source]
figure[figure omitted — 334 chars of source]
figure[figure omitted — 296 chars of source]

Results of the ATET for our simulation setting using a binary outcome are qualitatively similar to the continuous outcome. Table (ref) show the mean relative squared errors. Again, all figures are larger than 1, which means that using a non-optimal $k$ implies larger estimation errors.

table[table omitted — 2,294 chars of source]

Figure (ref) shows the relative 95% confidence intervals length ratios associated with $k$ and $k^*$. All six simulation settings look very similar with wider confidence intervals when $k$ is small, and narrower intervals when $k$ is large. There is stabilization around $k=10$. Figure (ref) shows type I errors using 5% as significance level. Larger $k$ implies larger errors, and using $k^*$ implies small type I errors in general, except when $k$ is very small. Small type I error can be explained by means of Figure (ref) where we see that optimal selection of $k$ is also based on small $k$. Here $k^*=\left\{2,3,4,5\right\}$ are particularly relevant.

In general, we can observe that our proposal implies relatively smaller estimation errors achieving a compromise between confidence interval length and type I error having as a result of small optimal $k$ most of the times.

figure[figure omitted — 415 chars of source]
figure[figure omitted — 337 chars of source]
figure[figure omitted — 300 chars of source]

Empirical example

To illustrate our proposal we estimate the average treatment effect on treated of 401(k) participation on accumulated net financial assets. 401(k) is a retirement plan where savings contributions are provided by employers deduced from employees' payment before taxes such that taxes on capital gains are avoided. Employees should achieve some eligibility criteria such that the plan does not have universal applicability in principle. The impact evaluation of 401(k) on net financial assets has been done previously by Benjamin2003,chernozhukov2004effects, Conley2012,chernozhukov2017double among others. We overcome the lack of random assignment following chernozhukov2017double arguments. Enrollment in a 401(k) plan is exogenous when controlling for observable variables which drive employment decisions, such as income, when the plan initially became available. Therefore, our dataset is from the 1991 Survey of Income and Program Participation composed by 9,915 observations.

We control for the same set of variables as chernozhukov2004effects. In particular, age (linear and squared), education (years), household size, income (seven levels), and binary variables indicating: defined benefit pension, home ownership, marital status (married), participation in IRA (individual retirement account) and two-earners status (both household heads contribute to household income). Details can be found in Benjamin2003.

Table (ref) shows unconditional mean difference tests of our outcome variables: accumulated net financial assets, and a binary variable indicating positive accumulated net financial assets. Employees enrolled in 401(k) plans have \$27.4 more accumulated net financial assets on average than employees who are not enrolled. The unconditional probability of positive net financial assets is 31% on average higher for the former. We can see that there are unconditional statistically significant differences.

table[table omitted — 1,258 chars of source]

Figure (ref) shows the ATET in the accumulated net financial assets using our optimal selection, $k^*=19$, and $k=1$, as a reference. We see similar ATET point estimates, which are approximately \$ 15K (similar outcomes are found by Conley2012 under the exogeneity assumption using a Bayesian approach for a local average treatment effects, LATE). However, the 95% confidence interval using $k^*$ is (\$10.9K,\$18.9K), whereas $k=1$ interval is (\$9.1K,\$19.9K), that is 35% larger. Results regarding the probability of positive accumulated net financial assets can be seen in Figure (ref). Both, $k^*=19$ and $k=1$ indicate statistically significant ATET whose point estimate is approximately 19.9% and 18.7% respectively. The 95% confidence intervals are (16.3%, 21.3%) and (17.7%, 21.3%) for $k=1$ and $k^*=19$. This means that using $k=1$ implies a 38.9% wider interval than using $k^*=19$. Observe that both exercises indicate the same optimal $k$. This means that diagnostic balance tests are based on the same control groups. Figure (ref) shows the love plot indicating that control groups based on $k^*=19$ satisfy mean balancing conditions.

figure[figure omitted — 468 chars of source]
figure[figure omitted — 500 chars of source]
figure[figure omitted — 341 chars of source]

Conclusions and final remarks

We present a simple approach to obtain an optimal selection of the number of control units ($k$) to build the “synthetic” or “counterfactual” unit for each individual in the k nearest neighbors algorithm. Our approach is based on a simple unbiased reference estimator (individual treatment effect on treated for ATET and individual treatment effect for ATE) such that we select $k$ minimizing the mean squared error with respect to the reference estimator. Therefore, the optimal $k=k^*$ achieves a balance between confidence interval length and type I error.

Our application suggests that $k^*$ can be relatively large compared to the common practice of using $k=1$. This implies significant reductions regarding confidence interval lengths. Although, both choices show 5% statistically significant effects of 401(k) enrollment in the probability of positive accumulated net financial assets, and its amount.

A final remark is that using as reference estimator for treatments effects the one obtained using $k=1$ is an interesting alternative. We obtained same qualitative results, that is, a balance between interval length and type I error, and a considerable reduction in MRSE. Simulations exercises confirm this.