Return Forecasts and Optimal Portfolio Construction: A Quantile Regression Approach

Ma, Lingjie; Pohlman, Larry

doi:10.2139/ssrn.880478

Cited by 9 publications

(8 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As Koenker and Bassett (1978) show, estimators of this type have high efficiency over a large class of distributions. A subset of the above specifications has been employed by Taylor (2007) and Ma and Pohlman (2008), among others.…”

Section: Forecasting Approaches Based On Quantile Regressionsmentioning

confidence: 99%

A Quantile Regression Approach to Equity Premium Prediction

Meligkotsidou

Panopoulou

Vrontos

et al. 2014

Journal of Forecasting

View full text Add to dashboard Cite

We propose a quantile regression approach to equity premium forecasting. Robust point forecasts are generated from a set of quantile forecasts, using both …xed and time-varying weighting schemes, thus exploiting the entire distributional information associated with each predictor. Further gains are achieved by incorporating the forecast combination methodology in our quantile regression setting. Our approach using a time-varying weighting scheme delivers statistically and economically signi…cant out-of-sample forecasts relative to both the historical average benchmark and the combined predictive mean regression modeling approach.JEL classi…cation: G11; G12; C22; C53

show abstract

Section: Forecasting Approaches Based On Quantile Regressionsmentioning

confidence: 99%

A Quantile Regression Approach to Equity Premium Prediction

Meligkotsidou

Panopoulou

Vrontos

et al. 2014

Journal of Forecasting

View full text Add to dashboard Cite

show abstract

“…Second, the conditional mean results could be derived from the conditional quantile effect. In particular, if the distribution of effects is not too skewed, the conditional mean effect would be close to the median (Ma & Pohlman 2008). In our analysis, for most variables, the conditional mean effects are higher than the conditional median effects, roughly between 70 th and 80 th quantiles, probably because of the skewness of our profit data.…”

Section: Model Construction and Interpretationmentioning

confidence: 57%

“…In the absence of quantile effects, the OLS model is a special case of quantile regression model (Ma & Pohlman 2008). All the regression quantile slope estimates are for a common parameter, and any deviation among the regression quantile estimates is simply due to sampling variation.…”

Section: Advantagesmentioning

confidence: 99%

“…Furthermore, changes in the shape of the distributions of y across the predictor variables can also be detected (Koenker & Machado 1999). In the condition of heterogeneity, quantile regression will provide a more complete view of the relationship between variables through the effects of independent variables across quantiles of the response distribution (Ma & Pohlman 2008). Potentially different solutions at distinct quantiles may be interpreted as differences in the response of the dependent variable to changes in the regressors at various points in the conditional distribution of the dependent variable (Buchinsky, 1998).…”

Section: Advantagesmentioning

confidence: 99%

“…However, since we are estimating many parameters, we must be cautious about our inferences. With more information, one has greater challenges in the decision-making (Ma & Pohlman 2008).…”

Section: Model Construction and Interpretationmentioning

confidence: 99%

See 2 more Smart Citations

Improving the profitability of direct marketing : a quantile regression approach

Zhang¹

View full text Add to dashboard Cite

Improving the Profitability of Direct Marketing:A Quantile Regression Approach by ZHANG Xi Master of PhilosophyDirect marketing is to target consumers who are most likely to respond. A number of target selection methods have been employed to select potential customers. These methods either only consider the customer response probability and ignore the profit issue or assume that the estimates of profit are homogenous across customers when considering the expected amount of profit. Furthermore, the traditional analytical techniques based on ordinary least squares (OLS) regression, which focus on the average customer, cannot examine the differences of various customer groups or account for customer heterogeneity in profitability estimates. Quantile regression, instead of the point estimate for the conditional mean, can be used to estimate the whole distribution, especially the upper tail which we are interested in. Quantile regression does not have strict model assumptions as OLS does and is not sensitive to outliers. To model consumer response profit in direct marketing, this thesis tested the endogeneity bias in the recency, frequency, monetary value (RFM) variables using the control function approach, made sample selection bias correction using Heckman's procedure, and then adopted quantile regression to estimate customer profit and make forecast of the profit distribution of future values. Furthermore, we adopted the recentered influence function (RIF) regression methods proposed by Firpo et al. (2007) to perform unconditional quantile regression for customer profit estimation. The comparison of OLS, conditional and unconditional quantile regression shows that while OLS may induce possible misleading estimation results, conditional and unconditional quantile regression can provide more informative estimation results. The findings can help direct marketers augment the profitability of marketing campaigns and have meaningful implications for solving target marketing forecasting problems given the constraint of limited resources. DECLARATION

show abstract

Predictable Return Distributions

Pedersen

2015

Journal of Forecasting

View full text Add to dashboard Cite

Using quantile regression this paper explores the predictability of the stock and bond return distributions as a function of economic state variables. The use of quantile regression allows us to examine specific parts of the return distribution such as the tails and the center, and for a sufficiently fine grid of quantiles we can trace out the entire distribution. A univariate quantile regression model is used to examine the marginal stock and bond return distributions, while a multivariate model is used to capture their joint distribution. An empirical analysis on US data shows that economic state variables predict the stock and bond return distributions in quite different ways in terms of, for example, location shifts, volatility and skewness. Comparing the different economic state variables in terms of their out-of-sample forecasting performance, the empirical analysis also shows that the relative accuracy of the state variables varies across the return distribution. Density forecasts based on an assumed normal distribution with forecasted mean and variance is compared to forecasts based on quantile estimates and, in general, the latter yields the best performance. Copyright

show abstract

Return Forecasts and Optimal Portfolio Construction: A Quantile Regression Approach

Cited by 9 publications

References 30 publications

A Quantile Regression Approach to Equity Premium Prediction

A Quantile Regression Approach to Equity Premium Prediction

Improving the profitability of direct marketing : a quantile regression approach

Predictable Return Distributions

Contact Info

Product

Resources

About