Machine learning for regularized survey forecast combination: Partially-egalitarian LASSO and its derivatives

Despite the clear success of forecast combination in many economic environments, several important issues remain incompletely resolved. The issues relate to the selection of the set of forecasts to combine, and whether some form of additional regularization (e.g., shrinkage) is desirable. Against this background, and also considering the frequently-found good performance of simple-average combinations, we propose a LASSO-based procedure that sets some combining weights to zero and shrinks the survivors toward equality (“partially-egalitarian LASSO”). Ex post analysis reveals that the optimal solution has a very simple form: the vast majority of forecasters should be discarded, and the remainder should be averaged. We therefore propose and explore direct subset-averaging procedures that are motivated by the structure of partially-egalitarian LASSO and the lessons learned, which, unlike LASSO, do not require the choice of a tuning parameter. Intriguingly, in an application to the European Central Bank Survey of Professional Forecasters, our procedures outperform simple average and median forecasts; indeed, they perform approximately as well as the ex post best forecaster.     

Macroeconomic data transformations matter

In a low-dimensional linear regression setup, considering linear transformations/combinations of predictors does not alter predictions. However, when the forecasting technology either uses shrinkage or is nonlinear, it does. This is precisely the fabric of the machine learning (ML) macroeconomic forecasting environment. Pre-processing of the data translates to an alteration of the regularization – explicit or implicit – embedded in ML algorithms. We review old transformations and propose new ones, then empirically evaluate their merits in a substantial pseudo-out-sample exercise. It is found that traditional factors should almost always be included as predictors and moving average rotations of the data can provide important gains for various forecasting targets. Also, we note that while predicting directly the average growth rate is equivalent to averaging separate horizon forecasts when using OLS-based techniques, the latter can substantially improve on the former when regularization and/or nonparametric nonlinearities are involved.     

Land regularization on agricultural frontiers: The case of Northwestern Petén, Guatemala

Land regularization, the provision of state-sanctioned property rights to landowners, is an important development strategy in the Global South. Although much work has examined the effects of regularization in settled rural areas, the effects of this policy on agricultural frontiers are poorly understood. Four benefits of regularization that are predicted for settled rural areas are improved tenure security, increased land-attached investment and credit access, and more efficient land market activity. This study explores the extent to which these benefits accrue to campesino landowners and society in the early years of a regularization project implemented by a non-government organization (NGO) on an agricultural frontier in Petén, Guatemala. Based on semi-structured interviews with landowners from three rural communities in northwestern Petén and discussions with personnel from state agencies and NGOs, this study concludes that the predicted benefits of regularization are strongly constrained by socio-economic and ecological factors in the agricultural frontier region, specifically, the weak connection between legal tenure and de facto tenure security, inadequate markets, high wildfire frequency, unwillingness of banks to supply campesino landowners with formal credit, and the presence of a strong extra-legal land market.     

A Survey of L1 Regression

L₁ regularization, or regularization with an L₁ penalty, is a popular idea in statistics and machine learning. This paper reviews the concept and application of L₁ regularization for regression. It is not our aim to present a comprehensive list of the utilities of the L₁ penalty in the regression setting. Rather, we focus on what we believe is the set of most representative uses of this regularization technique, which we describe in some detail. Thus, we deal with a number of L₁‐regularized methods for linear regression, generalized linear models, and time series analysis. Although this review targets practice rather than theory, we do give some theoretical details about L₁‐penalized linear regression, usually referred to as the least absolute shrinkage and selection operator (lasso).     

Smooth monotone covariance for elliptical distributions and applications in finance

Sample covariance is known to be a poor estimate when the data are scarce compared with the dimension. To reduce the estimation error, various structures are usually imposed on the covariance such as low-rank plus diagonal (factor models), banded models and sparse inverse covariances. We investigate a different non-parametric regularization method which assumes that the covariance is monotone and smooth. We study the smooth monotone covariance by analysing its performance in reducing various statistical distances and improving optimal portfolio selection. We also extend its use in non-Gaussian cases by incorporating various robust covariance estimates for elliptical distributions. Finally, we provide two empirical examples using Eurodollar futures and corporate bonds where the smooth monotone covariance improves the out-of-sample covariance prediction and portfolio optimization.     

Extra-legal land market dynamics on a Guatemalan agricultural frontier: Implications for neoliberal land policies

Neoliberal land policies such as land administration seek to improve property rights and the efficiency of land markets to boost rural economic production. Quantitative studies of pre-existing land markets can help planners to tailor these policies to local conditions. In this article we examine an extra-legal land market currently being modernized by a World Bank-sponsored land administration effort. Specifically, we use a hedonic-type revealed preference model and household survey data to estimate the factors affecting extra-legal land prices along an agricultural frontier in Petén, Guatemala. Our model indicates that land value is significantly affected by land attributes including location, tenure status, presence of water, distance to roads, and distance to landowners’ homes, and that land prices in the northwestern Petén are estimated to have risen on average 26.5% per year between 1977 and 2000. We contend that this rate of increase provides a strong incentive for colonists to speculate in land rather than invest in state sanctioned property rights. We conclude that if frontier development programs, such as land administration, are to become attractive to settlers in Petén and elsewhere, they must compete favorably with economic incentives associated with land speculation, or alternatively, target landowners who are not interested in playing the land market.     

Macroeconomic forecasting for Australia using a large number of predictors

A popular approach to forecasting macroeconomic variables is to utilize a large number of predictors. Several regularization and shrinkage methods can be used to exploit such high-dimensional datasets, and have been shown to improve forecast accuracy for the US economy. To assess whether similar results hold for economies with different characteristics, an Australian dataset containing observations on 151 aggregate and disaggregate economic series as well as 185 international variables, is introduced. An extensive empirical study is carried out investigating forecasts at different horizons, using a variety of methods and with information sets containing an increasing number of predictors. In contrast to other countries the results show that it is difficult to forecast Australian key macroeconomic variables more accurately than some simple benchmarks. In line with other studies we also find that there is little to no improvement in forecast accuracy when the number of predictors is expanded beyond 20–40 variables and international factors do not seem to help.     

抛物型扩散模型参数识别的正则化方法

经济计量模型参数识别是对观察的经济数据进行分析而得到经济结果的方法。从数学上讲属于反问题,而这类问题大多是不适定的。本文针对一类典型的抛物型扩散模型采用处理不适定问题的Tikhonov正则化方法来求解。即采用先离散化后正则化的策略利用在误差水平已知的情况下具有三阶收敛速率的算法来处理数值微分,并基于一种快速选择正则参数的混合算法计算正则解。同时对待辨识的分布参数线性和非线性依赖于观测数据两种情况分别进行了数值试验。     

A regularization approach to the many instruments problem

This paper focuses on the estimation of a finite dimensional parameter in a linear model where the number of instruments is very large or infinite. In order to improve the small sample properties of standard instrumental variable (IV) estimators, we propose three modified IV estimators based on three different ways of inverting the covariance matrix of the instruments. These inverses involve a regularization or smoothing parameter. It should be stressed that no restriction on the number of instruments is needed and that all the instruments are used in the estimation. We show that the three estimators are asymptotically normal and attain the semiparametric efficiency bound. Higher-order analysis of the MSE reveals that the bias of the modified estimators does not depend on the number of instruments. Finally, we suggest a data-driven method for selecting the regularization parameter. Interestingly, our regularization techniques lead to a consistent nonparametric estimation of the optimal instrument.     

地盘管采暖地面防裂控制措施

本文介绍了建筑工程中地盘管采暖地面在使用中出现裂缝的现象及形式、产生的机理和原因进行了研究分析，提出了防治措施和处理办法。