A Generalized Stepwise Procedure with Improved Power for Multiple Inequalities Testing

Hsu, Yu Chin; Kuan, Chung Ming; Yen, Meng Feng

doi:10.1093/jjfinec/nbu014

Cited by 24 publications

(20 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A lower bound p ‐value of 0.09 even indicates a statistically significant rejection of the null hypothesis at a 10% level when the SPA test is conducted in a more liberal way. Given that Hansen's SPA test is a very conservative test (e.g., Hsu, Kuan, & Yen, ), a look at the lower bound p ‐value seems quite sensible. In this respect, the VOL2‐12 model statistically significantly dominates the buy‐and‐hold benchmark.…”

Section: Simulation Resultsmentioning

confidence: 99%

“…As already mentioned, because of time and capacity limitations, within the group of fundamental forecast models and technical indicator models we only covered the (in our view) most popular strategies. However, the integration of further investment strategies into our test‐framework, as well as the additional application of a less conservative multiple test‐framework (e.g., Hsu et al, ), is straightforward and is left for future research.…”

Section: Discussionmentioning

confidence: 99%

“… In our longer backtest period ranging from 1966 to 2014 we observe a risk‐adjusted outperformance against the buy‐and‐hold‐strategy at a significance level around 10 percent. Given that Hansen's SPA test is a very conservative test (e.g., Hsu et al, ), this superiority could also be judged as significant. See the discussion in Section 6. …”

mentioning

confidence: 99%

See 2 more Smart Citations

Investing in the S&P 500 index: Can anything beat the buy‐and‐hold strategy?

Dichtl¹

2019

Rev Financ Econ

View full text Add to dashboard Cite

Determining whether investment strategies exist that provide higher (risk‐adjusted) returns than buying and holding the S&P 500 stock market index is not only highly relevant for finance theory, but also for the asset management industry. This study conducts a comprehensive test using realistic investment strategies based on monthly seasonalities, technical indicators, and fundamental factors (over 4,100 strategies in total). To assess statistical significance, we use Hansen's data‐snooping‐resistant SPA test. The results show that only investment strategies trying to exploit underreaction and overreaction effects with technical indicators dominate the buy‐and‐hold strategy in some simulation setups. These investment strategies are clearly superior to the strategies based on seasonalities and fundamental factors. Given that underreaction and overreaction effects are mainly justified with cognitive biases, our results support the economic relevance of behavioral finance insights.

show abstract

Section: Simulation Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Investing in the S&P 500 index: Can anything beat the buy‐and‐hold strategy?

Dichtl¹

2019

Rev Financ Econ

View full text Add to dashboard Cite

show abstract

“…Recent papers such as Hsu et al. () and Fishe and Smith () extend and apply existing false discovery proportion (FDP) and false discovery rate (FDR) controlling procedures to economics and finance settings. We provide formal definitions of various error rates in the Appendix and we refer the readers to Romano et al.…”

Section: Introductionmentioning

confidence: 99%

“…Hsu et al. () adopt the same idea in the context of FDP testing. Our tests do not impose the least favourable configuration either.…”

Section: Introductionmentioning

confidence: 99%

Oracle and adaptive false discovery rate controlling methods for one‐sided testing: theory and application in treatment effect evaluation

Shen

2017

The Econometrics Journal

View full text Add to dashboard Cite

Summary Economists are often interested in identifying effective policies or treatments together with subpopulations of individuals who respond positively (or with a sign that is expected) to these treatment interventions. In this paper, we propose an optimal false discovery rate controlling method that is especially useful for such one‐sided testing problems. The proposed procedure is optimal in the sense of minimizing the false non‐discovery rate while controlling the false discovery rate at a pre‐specified level; it uses a deconvolution method based on non‐parametric maximum likelihood estimation, which allows for a broader class of treatment effect distributions than existing methods do. The proposed test demonstrates good small‐sample performance in Monte Carlo simulations and it is applied to study the effect of attending a more selective high school in Romania. The application reveals strong evidence of treatment effect heterogeneity, in that students who marginally gain access to higher‐ranked schools are more likely to benefit if the higher‐ranked school has a relatively high admission score cut‐off – or, in other words, is more selective.

show abstract

Investment styles and the multiple testing of cross-sectional stock return predictability

Vincent

Hsu

Lin

2021

Journal of Financial Markets

View full text Add to dashboard Cite

A Generalized Stepwise Procedure with Improved Power for Multiple Inequalities Testing

Abstract: for very helpful discussions on this topic. We also thank Yi-Ting Yeh for helping simulations in the paper. C.-M.

Cited by 24 publications

References 37 publications

Investing in the S&P 500 index: Can anything beat the buy‐and‐hold strategy?

Investing in the S&P 500 index: Can anything beat the buy‐and‐hold strategy?

Oracle and adaptive false discovery rate controlling methods for one‐sided testing: theory and application in treatment effect evaluation

Investment styles and the multiple testing of cross-sectional stock return predictability

Contact Info

Product

Resources

About