Exact and Heuristic Approaches for the Index Tracking Problem with UCITS Constraints

Scozzari, Andrea; Tardella, Fabio; Paterlini, Sandra; Krink, Thiemo

doi:10.2139/ssrn.2066672

Cited by 5 publications

(8 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In the remaining papers, metaheuristics such as evolutionary algorithms (cf. [1,6,17,21,27,28,29]) or local-search heuristics (cf. [18,22,32]) are proposed.…”

Section: Related Literaturementioning

confidence: 99%

Tracking and outperforming large stock-market indices

Gnägi

Strub

2020

Omega

View full text Add to dashboard Cite

Highlights • We study the problem of tracking and outperforming large stock-market indices. • We compare linear and quadratic objective functions used in the literature. • We consider various real-life constraints that are relevant in practice. • We propose novel MIP formulations and novel matheuristics. • We find that the tracking error variance, a quadratic function, should be optimized.

show abstract

“…In the remaining papers, metaheuristics such as evolutionary algorithms (cf. [1,6,17,21,27,28,29]) or local-search heuristics (cf. [18,22,32]) are proposed.…”

Section: Related Literaturementioning

confidence: 99%

Tracking and outperforming large stock-market indices

Gnägi

Strub

2020

Omega

View full text Add to dashboard Cite

show abstract

“…Table 3 shows average tracking error results over all instances for α = 0.5 with L = −1 and L = 2, whilst varying the length of the in-sample period h between 13 and 52 weeks. It should be noted that, under column Num, the total number of rebalances for h = [13,26,39,52] is [30,29,28,27] respectively. The rows in this table for h = 52 are as the final rows in Tables 1 and 2, but are repeated here for ease of comparison.…”

Section: Varying H and αmentioning

confidence: 99%

“…Note here that since index tracking (so L = 1) has been extensively considered in the literature (e.g. see [3,10,11,13,16,24,29,30] for recent work) we focus here only on cases where L = 1, so portfolios of assets aiming to give a return different from that of the benchmark index.…”

mentioning

confidence: 99%

An optimisation approach to constructing an exchange-traded fund

2014

View full text Add to dashboard Cite

In this paper we consider the problem of deciding the portfolio of assets that should underlie an exchange-traded fund (ETF). We formulate this problem as a mixed-integer nonlinear program. We consider ETFs which have positive leverage with respect to their benchmark index and ETFs which have negative leverage (inverse, short, ETFs). Our formulation is a flexible one that incorporates decisions as to both long and short positions in assets, as well as including rebalancing and transaction cost. Computational results are given for problems, derived from universes defined by S&P international equity indices, involving up to 1,200 assets.

show abstract

“…O problema de index trackingé representadona literatura contemporânea com modelos que empregam diferentes métodos de solução. Analisando-se alguns desses artigos recentes, podemos agrupá-los por métodos de solução similares como, por exemplo, uso de métodos heurísticos (Beasley et al, 2003, Oh et al, 2005, Maringer & Oyewumi, 2007, Jeurissen & Van Den Berg, 2008, Krink et al, 2009, Guastaroba & Speranza, 2012, Scozzari et al, 2012, uso de cointegração (Dunis & Ho, 2005, Caldeira & Portugal, 2010 e uso de programação quadrática (Jansen & Van Dijk, 2002, Coleman et al, 2006, Yao et al, 2006. Gaivoronski et al (2005) discutem diferentes abordagens do problema de index tracking, realizando otimização apenas usando um solver e, também, aplicando métodos heurísticos.…”

Section: Revisão Da Literaturaunclassified

“…No caso de Beasley et al (2003), o objetivoé a formação de carteiras de 10 ativos com amostras de até 225 ativos -diferentemente do presente artigo, em que buscamos formar carteiras de tamanho superior com uma amostra menor. Maringer & Oyewumi (2007) e Scozzari et al (2012) também aplicam modelagens de programação quadrática e inteira, com controle da quantidade de ativos e também do peso máximo de ativos em cada carteira. Para solução, ambos os artigos usam heurísticas de evolução diferencial, devido ao uso de uma quantidade maior de restrições e o emprego de restrição inteira para formação de carteiras reduzidas.…”

Section: Revisão Da Literaturaunclassified

Index Tracking com Controle do Número de Ativos

2014

View full text Add to dashboard Cite

Index tracking is a passive investment strategy, which aims at generating portfolios to reproduce a specific market index’s performance. This article proposes a model for a index tracking problem with control on the number of assets in the portfolio, which corresponds to a restriction in transaction costs. The model is applied to Ibovespa (sample: 67 stocks) from January/2009 to July/2012. Portfolios were formed without limiting the amount of stocks and limiting this amount to 40, 30 and 20 stocks, with rebalancing periods of 20, 40 and 60 trading days. The results were satisfactory especially for the 60 days rebalancing period, in which transaction costs become lower due to the longer rebalancing period. We also verified that changes in Cplex parameters didn’t influence the results especially in relation to the computational times. Therefore, we also conclude about the need of using heuristic approaches to form portfolios with smaller amounts of assets.

show abstract

Exact and Heuristic Approaches for the Index Tracking Problem with UCITS Constraints

Cited by 5 publications

References 33 publications

Tracking and outperforming large stock-market indices

Tracking and outperforming large stock-market indices

An optimisation approach to constructing an exchange-traded fund

Index Tracking com Controle do Número de Ativos

Contact Info

Product

Resources

About