{"id":5205,"date":"2019-10-29T23:51:20","date_gmt":"2019-10-29T22:51:20","guid":{"rendered":"https:\/\/quantpedia.com\/?p=5205"},"modified":"2025-06-04T14:23:17","modified_gmt":"2025-06-04T12:23:17","slug":"quantpedia-premium-update-30th-october-2019","status":"publish","type":"post","link":"https:\/\/vvv.quantpedia.com\/es\/quantpedia-premium-update-30th-october-2019\/","title":{"rendered":"Quantpedia Premium Update – 30th October 2019"},"content":{"rendered":"

New strategies:<\/strong><\/p>\n\n\n\n

#453 – Machine Learning Adaptive Portfolio Asset Allocation<\/a><\/strong><\/p>\n\n\n\n

Period of rebalancing:<\/strong> Monthly
Markets traded: <\/strong>equities, bonds, REITs, commodities
Instruments used for trading:<\/strong> ETFs
Complexity:<\/strong> Very complex strategy
Backtest period:<\/strong> 2010 – 2017
Indicative performance:<\/strong> 10.07%
Estimated volatility:<\/strong> 10.13%<\/p>\n\n\n\n

Source paper:<\/strong><\/p>\n\n\n\n

Obeidat,Samer and Shapiro, Daniel and Lemay, Mathieu and MacPherson, Mary Kate and Bolic, Miodrag: Adaptive Portfolio Asset Allocation Optimization with Deep Learning<\/strong>
http:\/\/www.thinkmind.org\/download.php?articleid=intsys_v11_n12_2018_3<\/a>
Abstracto:
Portfolio management is a well-known multi-factor optimization problem facing investment advisors. The system described in this work can assist in automating portfolio management, and improving risk-adjusted returns. The asset allocation action recommendations were personalized to the portfolio under consideration, and were examined empirically in this work in comparison to standard portfolio management techniques. This work presents a Long Short-Term Memory approach to adaptive asset allocation, building upon prior work on training neural networks to model causality. The neural network model discussed in this work ingests historical price data and ingests macroeconomic data and market indicators using Principal Components Analysis. The model then estimates the expected return, volatility, and correlation for the selected assets. These neural network outputs were then turned into action recommendations using a Mean Variance Optimization framework augmented to use a forward looking rolling window technique. Testing was performed on a dataset with a 7.66 year duration. The observed mean annualized return for classical passive portfolio management approaches were 4.67%, 3.49%, and 4.57%, with mean Sharpe ratios of 0.46, 0.20, and 0.54. 10 simulations using the new Long Short Term Memory model from this work provided a mean annualized return of 10.07%, with a Sharpe ratio of 0.98. This work provides the conclusion that a Long Short-Term Memory model can generate better risk-adjusted returns than conventional strategic passive portfolio management. <\/p>\n\n\n\n

#454 – Time Series Momentum Strategies Using Deep Neural Networks<\/a><\/strong><\/p>\n\n\n\n

Period of rebalancing:<\/strong> Daily
Markets traded: <\/strong>commodities, bonds, equities, currencies
Instruments used for trading:<\/strong> futures, CFDs
Complexity:<\/strong> Very complex strategy
Backtest period:<\/strong> 1990 – 2015
Indicative performance:<\/strong> 14.10%
Estimated volatility:<\/strong> 15.40%<\/p>\n\n\n\n

Source paper:<\/strong><\/p>\n\n\n\n

Lim, Bryan and Zohren, Stefan and Roberts, Stephen: Enhancing Time Series Momentum Strategies <\/strong>
https:\/\/ssrn.com\/abstract=3369195 <\/a>
Abstracto:
While time series momentum is a well-studied phenomenon in finance, common strategies require the explicit definition of both a trend estimator and a position sizing rule. In this paper, we introduce Deep Momentum Networks — a hybrid approach which injects deep learning based trading rules into the volatility scaling framework of time series momentum. The model also simultaneously learns both trend estimation and position sizing in a data-driven manner, with networks directly trained by optimising the Sharpe ratio of the signal. Backtesting on a portfolio of 88 continuous futures contracts, we demonstrate that the Sharpe-optimised LSTM improved traditional methods by more than two times in the absence of transactions costs, and continue outperforming when considering transaction costs up to 2-3 basis points. To account for more illiquid assets, we also propose a turnover regularisation term which trains the network to factor in costs at run-time. <\/p>\n\n\n\n

#455 – Nonlinear Support Vector Machines and Stock Picking<\/a><\/strong><\/p>\n\n\n\n

Period of rebalancing:<\/strong> Quarterly
Markets traded: <\/strong>equities
Instruments used for trading:<\/strong> stocks
Complexity:<\/strong> Very complex strategy
Backtest period:<\/strong> 1981 – 2010
Indicative performance:<\/strong> 21.92%
Estimated volatility:<\/strong> 12.79%<\/p>\n\n\n\n

Source paper:<\/strong><\/p>\n\n\n\n

Huerta, Ramon, Corbachob, Fernando and Elkan, Charles : Nonlinear support vector machines can systematically identify stocks with high and low future returns<\/strong>
https:\/\/biocircuits.ucsd.edu\/huerta\/AF_Huerta.pdf<\/a><\/a>
Abstracto:
This paper investigates the pro\ufb01tability of a trading strategy based on training a model to identify stocks with high or low predicted returns. A tail set is de\ufb01ned to be a group of stocks whose volatility-adjusted price change is in the highest or lowest quantile, for example the highest or lowest 5%. Each stock is represented by a set of technical and fundamental features computed using CRSP and Compustat data. A classi\ufb01er is trained on historical tail sets and tested on future data. The classi\ufb01er is chosen to be a nonlinear support vector machine (SVM) due to its simplicity and effectiveness. The SVM is trained once per month, in order to adjust to changing market conditions. Portfolios are formed by ranking stocks using the classi\ufb01er output. The highest ranked stocks are used for long positions and the lowest ranked ones for short sales. The Global Industry Classi\ufb01cation Standard is used to build a model for each sector such that a total of 8 long-short portfolios for Energy, Materials, Industrials, Consumer Discretionary, Consumer Staples, Health Care, Financials, and Information Technology are formed. The data range from 1981 to 2010. Without measuring trading costs, but using 91 day holding periods to minimize these, the strategy leads to annual excess returns (Jensen alpha) of 15% with volatilities under 8% using the top 25% of the stocks of the distribution for training long positions and the bottom 25% for the short ones.<\/p>\n\n\n\n

New research papers related to existing strategies:<\/strong><\/p>\n\n\n\n

#73 – Pairs Trading with Commodities<\/a><\/strong><\/p>\n\n\n\n

Mooney, Rapaka, Vera: Dynamic Regime Strategy for Stress Testing and Optimizing Institutional Investor Portfolios<\/strong>
https:\/\/papers.ssrn.com\/sol3\/papers.cfm?abstract_id=3438272<\/a>
Abstracto:
Our work aims to develop a stand-alone trading system to construct portfolios that show the benefits of value and momentum style integration and presents the effectiveness of alternative integration methods for long-only absolute return funds, which seeks absolute returns that are not highly correlated to traditional assets such as stocks and bonds. Our approach uses the CRoss Industry Standard Process for Data Mining (CRISP-DM) model to guide the necessary steps, processes, and workflows for executing our project. <\/p>\n\n\n\n

#152 – Momentum Factor Effect in REITs<\/a>
#159 – Value Effect in REITs<\/a><\/strong><\/p>\n\n\n\n

Guidolin, Pedio: How Smart Is the Real Estate Smart Beta? Evidence from Optimal Style Factor Strategies for REITs<\/strong>
https:\/\/papers.ssrn.com\/sol3\/papers.cfm?abstract_id=3458308<\/a>
Abstracto:
This paper has a twofold objective. First, we contribute to the stream of literature that investigates whether traditional asset pricing factors show any predictive power for the cross-section of Real Estate Investment Trust (REIT) returns. In particular, we investigate the existence of a premium associated to the Value, Size, Momentum, Investment, and Profitability factors over the period 1993-2018. We find support for all the pricing factors but for the Profitability one. Second, we investigate whether a set of smart beta strategies, based on the combination of the identified factors, may outperform similar allocation techniques that do not exploit factors. We find that all the proposed factor-based strategies display a higher risk-adjusted out-of-sample performance than a simple buy-and-hold investment in the real estate market (proxied by the FTSE NAREIT All REITs Index). In addition, we find that when factor-based strategies are implemented, REIT-only portfolios display risk-adjusted performances comparable to those of diversified portfolios that include equity, bond, and commodities. <\/p>\n\n\n\n

And three short free blog posts have been published during last 2 weeks:<\/strong><\/p>\n\n\n\n

Two blogs related to an interesting financial research papers:<\/strong><\/p>\n\n\n\n

Do you want to have an outperforming hedge fund? Then write a description of your investment strategy more creatively, clearly and use more synonyms\u2026 Of course, I am just kidding. However, a recent academic study written by Joenv\u00e4\u00e4r\u00e4, Karppinen, Teo, and Tiu shows that text sophistication can be used to find skilled hedge fund managers. Lexical diversity is the propensity of the writer to use multiple synonyms rather than repeated words. Skilled and, therefore, cognitively gifted managers are more likely to use rich vocabulary when writing their strategy descriptions. Therefore, if you feel that your favorite manager composes clear and captivating texts, maybe he is skilled also in his primary role \u2013 investment management \u2026<\/strong><\/p>\n\n\n\n

Joenv\u00e4\u00e4r\u00e4, Karppinen, Teo, Tiu: Text Sophistication and Sophisticated Investors<\/strong>
https:\/\/papers.ssrn.com\/sol3\/papers.cfm?abstract_id=3438758<\/a>
Abstracto:
We show that two novel measures of text sophistication, applied to hedge fund strategy descriptions, encapsulate incremental information about funds. Consistent with the linguistics literature, hedge funds with lexically diverse strategy descriptions outperform, eschew tail risk, and encounter fewer regulatory problems. In line with the literature, hedge funds with syntactically complex strategy descriptions report more regulatory violations and trigger more severe infractions. Fund investors recognize the dichotomy and direct flows accordingly, but not enough to erode away the alphas of lexically diverse funds. Our findings suggest that text sophistication measures provide texture on the cognitive ability and trustworthiness of sophisticated investors<\/p>\n\n\n\n

We at Quantpedia absolutely love long-term studies, and academic\n research paper written by Bhardwaj, Janardanan, and Rouwenhorst is \nreally exceptional. There are a lot of studies covering a long history \nof equity and bond markets. But futures markets are not covered so well,\n and that\u2019s the reason why is this paper so valuable. An additional plus\n is that study covers also delisted contracts, which makes the study\u2019s \ndata quality even better. Quantpedia\u2019s recommended read to anyone \ninterested in asset allocation into commodities \u2026<\/strong><\/p>\n\n\n\n

Bhardwaj, Janardanan and Rouwenhorst: The Commodity Futures Risk Premium: 1871\u20132018<\/strong>
https:\/\/papers.ssrn.com\/sol3\/papers.cfm?abstract_id=3452255<\/a>
Abstracto:
Using a novel comprehensive database of 230 commodity futures that traded between 1871 and 2018, we document that futures prices have on average been set at a discount to future spot prices by about 5%. The historical risk premium is robust across commodity sectors and varies with the state of the economy, inflation and the level of scarcity. Although the majority of contracts are defunct, most commodities have earned a positive risk premium over their lifespan. We find empirical support for Gray\u2019s conjecture that survival of futures contracts is correlated with the returns earned by investors. Finally, we provide out-of-sample evidence that \u201cfactor\u201d strategies based on commodity basis and momentum have historically earned positive returns over time but are subject to prolonged drawdowns (crashes) that are not dissimilar to those experienced by the overall market. <\/p>\n\n\n\n

Plus one short Quantpedia analysis:<\/strong><\/p>\n\n\n\n

<\/p>\n\n\n\n

We are continuing in our short series of articles about calendar \/ seasonal trading. The main focus of this paper is to show that the well-working calendar \/ seasonal anomalies can be refined. The aim is to find the right factors and find a way how to combine them in a search for profit from the practitioner\u2019s point of view. Based on our previous research, calendar anomalies are profitable, but there is a possible way how to enhance their performance. This can be done by employing momentum strategies. By assigning a weight to assets from a diversified set according to their momentum value, it is possible to find a profitable asset during various global market conditions. Moreover, a trend factor is used to ensure that when market conditions are not favorable, the strategy will not trade. Such addition is a typical approach used for reducing maximal draw-downs. Finally, since this paper is written from the practitioner\u2019s point of view, we are assuming some model transaction costs and examine the strategy in their presence.<\/strong><\/p>\n\n\n\n

Vojtko, Padysak: Calendar \/ Seasonal Trading and Momentum Factor
https:\/\/\\\/\\\/new-fmhwbzh6ghd9hede.swedencentral-01.azurewebsites.net\/calendar-seasonal-trading-and-momentum-factor\/<\/a><\/p>\n\n\n\n

\n\n\n\n

\u00bfBuscas m\u00e1s estrategias para leer? Suscr\u00edbete a nuestro bolet\u00edn informativo<\/a> o visite nuestra Blog<\/a> o Evaluador<\/a><\/strong>.<\/p>\n\n\n\n

\u00bfQuieres saber m\u00e1s sobre el servicio Quantpedia Premium? Consulta C\u00f3mo funciona Quantpedia<\/a>, nuestra misi\u00f3n<\/a> y Oferta de precios premium<\/a>.<\/strong><\/p>\n\n\n\n

\u00bfQuieres saber m\u00e1s sobre el servicio Quantpedia Pro? Compru\u00e9balo descripci\u00f3n<\/a>, mirar videos<\/a>, revisar capacidades de generaci\u00f3n de informes<\/a> y visite nuestro oferta de precios<\/a>.<\/strong><\/p>\n\n\n\n

\u00bfBuscas datos hist\u00f3ricos o plataformas de backtesting? Consulta nuestra lista de Descuentos en Algo Trading<\/a><\/strong>.<\/p>\n\n\n\n

\u00bfTe gustar\u00eda tener acceso gratuito a? nuestros servicios<\/a>? Entonces, Abre una cuenta con Lightspeed.<\/a> y disfrute de un a\u00f1o de Quantpedia Premium sin costo alguno.<\/strong><\/p>\n\n\n\n

\n\n\n\n

O s\u00edguenos en:<\/strong><\/p>\n\n\n\n

Facebook Grupo<\/a>, Facebook P\u00e1gina<\/a>, Gorjeo<\/a>, LinkedIn<\/a>, Medio<\/a> o YouTube<\/a><\/strong><\/p>","protected":false},"excerpt":{"rendered":"

Three new strategies have been added:<\/p>\n

Two new related research papers have been included into existing strategy reviews. And three short free blog posts<\/strong><\/a> about interesting related research papers have been published during last few weeks.<\/p>","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-5205","post","type-post","status-publish","format-standard","hentry","category-uncategorized"],"_links":{"self":[{"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/posts\/5205","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/comments?post=5205"}],"version-history":[{"count":0,"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/posts\/5205\/revisions"}],"wp:attachment":[{"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/media?parent=5205"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/categories?post=5205"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/vvv.quantpedia.com\/es\/wp-json\/wp\/v2\/tags?post=5205"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}