Similar Portfolios Research Articles

Designing and deploying insurance recommender systems using machine learning

AbstractRecommender systems have become extremely important to various types of industries where customer interaction and feedback is paramount to the success of the business. For companies that face changes that arise with ever‐growing markets, providing product recommendations to new and existing customers is a challenge. Our goal is to give our customers personalized recommendations based on what other similar people with similar portfolios have, in order to make sure they are adequately covered for their needs. Our system uses customer characteristics in addition to customer portfolio data. Since the number of possible recommendable products is relatively small, compared to other recommender domains, and missing data is relatively frequent, we chose to use Bayesian Networks for modeling our systems. We also present a deep‐learning‐based approach to provide recommendations to prospects (potential customers) where only external marketing data is available at the time of prediction.This article is categorized under: Application Areas > Industry Specific Applications Algorithmic Development > Structure Discovery Algorithmic Development > Bayesian Models Technologies > Machine Learning

Engel\u2019s law in the commodity composition of exports

Different shares of distinct commodity sectors in production, trade, and consumption illustrate how resources and capital are allocated and invested. Economic progress has been claimed to change the share distribution in a universal manner as exemplified by the Engel’s law for the household expenditure and the shift from primary to manufacturing and service sector in the three sector model. Searching for large-scale quantitative evidence of such correlation, we analyze the gross-domestic product (GDP) and international trade data based on the standard international trade classification (SITC) in the period 1962 to 2000. Three categories, among ten in the SITC, are found to have their export shares significantly correlated with the GDP over countries and time; The machinery category has positive and food and crude materials have negative correlations. The export shares of commodity categories of a country are related to its GDP by a power-law with the exponents characterizing the GDP-elasticity of their export shares. The distance between two countries in terms of their export portfolios is measured to identify several clusters of countries sharing similar portfolios in 1962 and 2000. We show that the countries whose GDP is increased significantly in the period are likely to transit to the clusters displaying large share of the machinery category.

Alumni Networks and Investment Strategy: Evidence from Chinese Mutual Funds

ABSTRACT While fund managers are connected by interpersonal linkages, the investment strategy of a mutual fund would follow a similar pattern due to interpersonal communications and information exchange. This paper focuses on the connection of fund managers via a shared educational network (university alumni connection) in China. Based on a manually collected dataset with the detailed transaction and manager information, our results were threefold: (1) Mutual fund managers with an alumni connection tend to have similar portfolio allocation; the network effect is heterogeneous for different manager characteristics, fund characteristics, and alumni types. (2) A dynamic learning strategy for fund managers exists where they tend to mimic the past portfolio choices of others within the alumni network. (3) The performance of connected funds is worsened by the degree of the connection.

Market timing skill of foreign portfolio investors in India

This study examines the market timing skill of foreign portfolio investors (foreign institutional investors or FIIs) in India as a group. A simulated portfolio is constructed that invests in the market index replicating the patterns of aggregate net FII flow. Similar portfolios are constructed replicating the trading behaviour of domestic institutional investors (DIIs) and investors of other categories. The performance of these portfolios is compared with the performance of two benchmark portfolios. The results based on bootstrapped modified internal rate of return of investment portfolios suggest that the FII portfolio does not outperform the benchmark portfolios, nor the portfolios representing investment strategies of DIIs and other investors. Our results suggest that the FIIs do not have superior market timing skills.

Managers beyond borders: side-by-side management in mutual funds and pension funds

Firms of collective investment industries, such as mutual and pension funds, are knowledge-intensive organizations that share managers between their multi-businesses to transfer the know-how of their employees and take advantage of the firm-specific human capital. This paper analyses the determinants and consequences of the simultaneous management (also called side-by-side or SBS management) of mutual and pension funds. In a sample of UK domestic equity mutual and pension funds, we find that managers are more likely to become SBS managers whether they belong to larger firms, have prior experience handling multiple funds, possess higher tenure, are female, and run smaller funds. Furthermore, SBS managers outperform their peers, showing the value of sharing human resources among diversified firms. Our results also show that holding similar portfolios between the funds of a manager may facilitate the multi-fund management. Top performers (SBS and non-SBS) present higher portfolio similarities than bottom performers, and higher fund similarities allow top managers to attract flows between their funds. However, top non-SBS managers erode performance due to higher portfolio similarities.

Diversification and Systemic Bank Panics

Diversification through pooling and tranching securities was supposed to mitigate creditor runs in financial institutions by reducing their credit risk, yet many financial institutions holding diversified portfolios experienced creditor runs in the recent financial crisis of 2007-2009. We present a theoretical model to explain this puzzle. In our model, because financial institutions all hold similar (diversified) portfolios, their behavior in the asset market is clustered: they either sell their assets at the same time or collectively do not sell. Such clustering behavior reduces market liquidity after an adverse shock and increases the probability of a panic run by creditors. We show that diversification, while making the financial system more robust against small shocks, increases the possibility of a systemic crisis in the case of a larger shock; diversification, inducing stronger strategic complementarities across institutions, makes a self-fulfilling systemic crisis (multiple equilibria) more likely. Because individual institutions either over-diversify or under-diversity in the competitive equilibrium compared with the social optimum, there is room for regulation.

Portfolio Similarity and Asset Liquidation in the Insurance Industry

An important assumption underlying the designation of some insurers as systemically important is that their overlapping portfolio holdings can result in common selling. We measure the overlap in holdings using cosine similarity, and show that insurers with more similar portfolios have larger subsequent common sales. This relationship can be magnified for some insurers when they are regulatory capital constrained or markets are under stress. When faced with an exogenous liquidity shock, insurers with greater portfolio similarity have even larger common sales that impact prices. Our measure can be used by regulators to predict which institutions may contribute most to financial instability through the asset liquidation channel of risk transmission.

Normative portfolio theory

In this paper, we correct the adverse impact of estimation risk on both portfolio weights and performance with two new equity allocation methods we implement with estimation-free and estimated ex-ante returns. Portfolios with estimation-free ex-ante returns and systematic-to-unsystematic risk weights have statistically higher Sharpe ratios than both similar portfolios with estimated ex-ante returns and 1/N′th portfolios. Optimal portfolio methods with well-behaved weights guide investors in a way not hitherto possible (normative portfolio theory).

Risk measures in a quantile regression credibility framework with Fama/French data applications

In this paper we extend the idea of embedding the classical credibility model into risk measures, as was presented by Pitselis (2016), to the idea of embedding regression credibility into risk measures. The resulting credible regression risk measures capture the risk of individual insurer’s contract (in finance, the individual asset return portfolio) as well as the portfolio risk consisting of several similar but not identical contracts (in finance, several similar portfolios of asset returns), which are grouped together to share the risk. In insurance, credibility plays a special role of spreading the risk. In financial terminology, credibility plays a special role of diversification of risk. For each model, regression credibility models are established and the robustness of these models is investigated. Applications to Fama/French financial portfolio data are also presented.

Fracking, Drilling, and Asset Pricing: Estimating the Economic Benefits of the Shale Revolution

We quantify the effect of a significant technological innovation, shale oil development, on asset prices. Using stock returns on major news announcement days allows us to link aggregate stock price fluctuations to shale technology innovations. We exploit cross-sectional variation in industry portfolio returns on days of major shale oil-related news announcements to construct a shale mimicking portfolio. This portfolio can explain a significant amount of variation in aggregate stock market returns, but only during the time period of shale oil development, which begins in 2012. Our estimates imply that $3.5 trillion of the increase in aggregate U.S. equity market capitalization since 2012 can be explained by this mimicking portfolio. Similar portfolios based on major monetary policy announcements do not explain the positive market returns over this period. We also show that exposure to shale oil technology has significant explanatory power for the cross-section of employment growth rates of U.S. industries over this period.

Multi-period portfolio optimization: Translation of autocorrelation risk to excess variance

Growth-optimal portfolios are guaranteed to accumulate higher wealth than any other investment strategy in the long run. However, they tend to be risky in the short term. For serially uncorrelated markets, similar portfolios with more robust guarantees have been recently proposed. This paper extends these robust portfolios by accommodating non-zero autocorrelations that may reflect investors’ beliefs about market movements. Moreover, we prove that the risk incurred by such autocorrelations can be absorbed by modifying the covariance matrix of asset returns.

Do Politicians “Put Their Money Where Their Mouth Is?” Ideology and Portfolio Choice

We examine the role of political ideology in portfolio formation by studying a unique set of investors whose ideology can be precisely captured by a well-defined, continuous measure and whose personal asset allocation decisions are mandatorily disclosed, namely, the members of the U.S. Congress. As such, we overcome important methodological issues facing previous work in this area. We find that politicians with similar beliefs hold similar portfolios and that more liberal members engage in more socially responsible investing (SRI), even within political parties. Politicians disproportionately favor the SRI categories that reflect their favorite issues, while salience plays an important role in activating their ideologically based preferences for SRI. In addition, more ideological investors are less likely to engage in quid pro quo behavior. We conclude that ideology is a pervasive psychological factor that governs decisions across the domains of politics, investing, and, even, ethics. This paper was accepted by Lauren Cohen, finance.

Network Ties and Organizational Action: Explaining Variation in Social Service Provision Patterns

This study integrates social capital theory and network analysis to explore the relationship betweeninterorganizational networks and organizational action. It analyzes the collaborative partnerships that religiouscongregations form to provide social services, and it examines how these ties are associated with the number andtypes of programs they offer. Using cross-sectional and panel data from a national study of congregations, theanalysis finds significant relationships between congregations’ interorganizational ties and their social serviceprovision patterns. Congregations that collaborate with other organizations offer more programs, and the effect iseven greater for congregations with a diverse portfolio of collaborators. Furthermore, a network analysis indicatesthat congregations with a similar portfolio of collaborators offer a similar menu of services. This study demonstratesthat an organization’s collaborative ties, above and beyond its internal characteristics, are significantly associatedwith both the volume and scope of its activity.

Forest planning and productivity-risk trade-off through the Markowitz mean-variance model

In order to make a comparative assessment between productivity and risk, we study the forest planning by means of the Markowitz mean-value (M-V) portfolio model. By weighting the forest productivity with factors of future climate change effects, we compute the optimal tree species mixes, within reach of forest managers, in ninety French administrative departments. Considering three different productivity measures (wood production, carbon sequestration and economic valorization) and their respective variances, we find that: a) the empirical allocation lies between the optimizations of wood production and economic valorization; b) forest managers prefer low variance to high productivity, i.e. their revealed risk aversion is high; and c) unlike maximizing wood productivity or carbon sequestration, which leads to similar portfolios, maximizing the economic value of wood production decreases both the levels of wood production and carbon sequestration. Under high risk aversion, the economic valorization would lead to a high species specialization, which is very unlikely in reality. In all considered scenarios, the objectives set out in the Kyoto Protocol would be attained, which puts into question its relevance in terms of additionality.

Do Mutual Fund Investors Get What They Pay For? The Legal Consequences of Closet Index Funds

Actively managed mutual funds sell the potential to beat the market by picking stocks that are expected to outperform passive benchmarks like the S&P 500. Funds that are marketed as active vary substantially in the degree to which their portfolio holdings actually differ from the holdings of passive index funds. A purportedly active fund with a portfolio that substantially overlaps with the market is known as a closet index fund. Since closet index funds charge considerably higher fees than true index funds but provide a substantially similar portfolio, they tend to be poor investment choices. This article presents empirical evidence on closet index funds, showing that more than 10% of U.S. mutual fund assets are currently invested in closet index funds and that high cost closet index funds substantially underperform their benchmarks. We argue that persistent closet indexing implicates a number of legal issues, including possible liability for fund advisors under the Securities Act and the Investment Company Act. We conclude by discussing potential adjustments to mutual fund disclosures that could help investors identify closet index funds.

Why an open access publishing cooperative can work

Why an open access publishing cooperative can work

Stock portfolio structure of individual investors infers future trading behavior.

Although the understanding of and motivation behind individual trading behavior is an important puzzle in finance, little is known about the connection between an investor's portfolio structure and her trading behavior in practice. In this paper, we investigate the relation between what stocks investors hold, and what stocks they buy, and show that investors with similar portfolio structures to a great extent trade in a similar way. With data from the central register of shareholdings in Sweden, we model the market in a similarity network, by considering investors as nodes, connected with links representing portfolio similarity. From the network, we find investor groups that not only identify different investment strategies, but also represent individual investors trading in a similar way. These findings suggest that the stock portfolios of investors hold meaningful information, which could be used to earn a better understanding of stock market dynamics.

Inter- and Intraobserver Agreement in Scoring Angiographic Results of Intra-Arterial Stroke Therapy

Angiographic results are commonly used as surrogate markers of the success of intra-arterial therapies for acute stroke. Inter- and intraobserver agreement in judging angiographic results remain poorly characterized. Our goal was to assess 2 commonly used revascularization scales. A portfolio of 148 pre- and post treatment images of 37 cases of proximal anterior circulation occlusions was electronically sent to 12 expert observers who were asked to grade treatment outcomes according to recanalization (of arterial occlusive lesion) or reperfusion (TICI) scales. Three expert observers had to score treatment outcomes by using a similar portfolio of 32 patients or when they had full access to all angiographic data, twice for each method 3-12 months apart. Results were analyzed by using κ statistics. Agreement among 9 responding observers was moderate for both the TICI (κ = 0.45 ± 0.01) and arterial occlusive lesion (κ = 0.39 ± 0.16) scales. Agreement was similar (moderate) when 3 observers had access to a portfolio (κ = 0.59 ± 0.06 and 0.49 ± 0.07, respectively) or to the full angiographic data (κ = 0.54 ± 0.06 and 0.59 ± 0.07, respectively). Intraobserver agreement was "fair to moderate" for both methods. Interobserver agreement became "substantial" (>0.6) when outcomes were dichotomized into "success" (TICI 2b, 3; arterial occlusive lesion II, III or "failure"; the results were judged more favorably when the arterial occlusive lesion rather than the TICI scale was used. There is an important variability in the assessment of angiographic outcomes of endovascular treatments, invalidating comparisons among publications. A simple dichotomous judgment can be used as a surrogate outcome when treatments are assessed by the same observers in randomized trials.

Can Linear Predictability Models Time Bull and Bear Real Estate Markets? Out-of-Sample Evidence from REIT Portfolios

A recent literature has shown that REIT returns contain strong evidence of bull and bear dynamic regimes that may be best captured using nonlinear econometric models of the Markov switching type. In fact, REIT returns would display regime shifts that are more abrupt and persistent than in the case of other asset classes. In this paper we ask whether and how simple linear predictability models of the vector autoregressive (VAR) type may be extended to capture the bull and bear patterns typical of many asset classes, including REITs. We find that nonlinearities are so deep that it is impossibile for a large family of VAR models to either produce similar portfolio weights or to yield realized, ex-post out-of-sample long-horizon portfolio performances that may compete with those typical of bull and bear models. A typical investor with intermediate risk aversion and a 5-year horizon ought to be ready to pay an annual fee of up to 5.7 % to have access to forecasts of REIT returns that take their bull and bear dynamics into account instead of simpler, linear forecast.

Attention Networks: A Two-Mode Network View on Valuation

When multiple actors allocate their attention across multiple issues, they create an attention network. We leverage the multiple ties that comprise such an attention network to argue that how competitors interpret a given market situation depends not only on information about that situation but also on the portfolio of other situations to which they pay attention. Specifically, we hypothesize that the more two competitors assess a given situation vis-a-vis a more similar portfolio of other situations, the more they assess that issue similarly. It is so, we argue, because i) competitors who pay attention to the same problems are more likely to make similar cognitive associations and consequently come to more similar solutions on a given problem, and ii) competitors are more likely to adapt their interpretations on the basis of the assessments made by the competitors with whom their attention patterns intersect more frequently. We exploit the two-mode (agents-issues) dynamic structure of these observational networks to study the social and temporal aspects of financial analysts’ valuations in markets from 1993-2011. Our findings show that two analysts form more similar earnings estimates about a given firm if they come to that assessment while paying attention to a more similar portfolio of other firms. Our analysis further demonstrates that analysts’ estimates of the focal firm are influenced by views that cycle through closed triads.