Advanced Portfolio and Investment Theory

Transcription

1 Advanced Portfolio and Investment Theory Week of January 28, 2013 Introduction & Portfolio Management 1.1 Principals David R Audley, Ph.D.; Sr. Lecturer in AMS david.audley@jhu.edu Office: WH 212-A ( ) Hours: Monday 4:30pm 6:00pm Teaching Assistant(s) Cheng, Wan-Schwin (Allen), (wcheng14@jhu.edu) Office Hours: TBD Wang, Heng (hwang82@jhu.edu) Office Hours: TBD Computing TA: Sancar Adali (211-D) sadali1@jhu.edu 1.2 Schedule Resources Lecture Encounters 3:00 PM in Whitehead 304 (to 4:15pm) Section 4:30 PM in Whitehead 304 (to 5:20 pm) No Section Jan 29. Textbooks Elton, Gruber, Brown, and Goetzmann: Modern PF Theory and Investment Analysis (8e), John Wiley, 2010 Noel Amenc & Veronique Le Sourd: Portfolio Theory and Performance Analysis, John Wiley 2003 Supplemental TBA On Reserve in Library

2 Resources Course Home Page: AMS Website TBA Additional Subject Material Class Resources & Lecture Slides Academic, Industry & Street Research (Optional) Consult at your leisure/risk or as assigned Interest can generate Special Topics sessions Measures of Performance Mid Term Exam (1/3 of grade) Final Exam (1/3 of grade) Homework (as assigned) and Pop Quizzes (1/3 of grade) Includes Class Participation and Attendance Un-excused Absence for Quiz results in zero credit toward that average Problem sets will be due each Monday, when one is assigned Assignment For January 28 (This Week) Read: A&L, Introduction & Chapter 1 (PF Management ) For February 4 (Next Week) Read: A&L, Chapter 1 (continued) Chapter 2 (Basic Performance Analysis Concepts) Section 1 (Return Calculations) Introduction Portfolio/Asset Management Past 25 years and earlier Graham & Dodd (1934), Security Analysis Homer & Liebowitz (1972), Inside the Yield Book Fidelity, Salomon, Lehman BlackRock, PIMCO (Allianz) From an Art & Craft to an Industry A rigorous analytic science Integrated Business vs. Multi Management (FofF)

3 Introduction Portfolio/Asset Management (the trend) Production: Multi-management Manager & Asset Class Specialization Financial Diversification (asset classes & style) Organizational Diversification (managers) Distribution: 3 rd Party, Open Architecture Has influenced financial thinking Professional & Academic Theory & Innovation Introduction Portfolio/Asset Management (the trend) Multi-Management & Multi-Distribution Delegate to the best managers Sell the value-add of manager selection & allocation Specialization by Managers Value/Risk analysis tools and hedging Performance analysis and enhancement Marketing: Academic evidence & scientific proof of rigor and superiority (?!?!) Introduction 447 / Amenc & Le Sourd Place practice, empirical studies, & innovations in context All are not major, paradigm-changing as is often suggested by promoters Separate true innovations from mere improvements to the well-known & existing Situate the importance of innovation w/r to the Fundamental Portfolio Management Questions 1.11 Introduction 447 / Amenc & Le Sourd Fundamental Portfolio Management Questions: Evolution of the investment management process Risk analysis, measurement and control Performance analysis, measurement and enhancement Identify and Account for explicit/implicit assumptions in the promoted tool/technique and evaluate inherent interpretive or practical limits

4 Categories of Assets Asset Classes usually classed by nature and/or level of risk Equities Bonds Money Market (Cash) Derivatives: Futures, Forwards, Swaps, Options, Alternatives: Hedge Funds, Managed Futures, Commodities, Private Equity, Venture, Private Debt, Real Estate, Collectables, 1.13 Categories of Assets Grouping by Sector Most common in Equities, but becoming more so in other classes as well Equity Sectors Automobiles Banks Primary Products Chemicals Construction Goods & Svcs Energy Financial Services Food Processing Pharma Distribution Tech Telecomm Public Services Health Care Capitalization (big/small/micro/...) Growth/Value Specialization of Funds: A trend, by sector, Cap A portfolio is defined as a grouping of assets Constructing the PF Making it Evolve to reach the Return Objective defined by the investor, and Respect the Investor s Constraint on Risk and Asset Allocation Investment methods to reach the Objective Quantitative Investment (origins in MPT) Most Widely Used Traditional Methods in Financial Analysis Quantitative Methods Active Management Passive Management A Difficult Separation

5 Passive Investment Management A Tracking Strategy (usually, the market/index) No Attempt to Anticipate the Market s Evolution Rationalizing Assumption Markets are perfectly efficient; fruitless to beat the market Therefore, the best technique is to try and replicate it Index funds that replicated the market/index have the lowest fees 1.17 Passive Investment Management Tilted Index funds Introduce an element of Active Management (an overlay ) Seek performance better than index Avoid market risk greater than index Performance differential w/reference portfolio: tracking error Goal: beat index regularly, not by a significant amount Tracking error: measured w/precision & kept strictly in-band Technique: analyze systematic PF risk w/multifactor model, exploit most lucrative risk factors (stock picking, IR, credit, ) 1.18 Passive Investment Management PF Insurance Adjust mix between low risk (e.g., money market) and risky assets so return never falls below a certain level Between two asset classes stocks and bonds Tactical Asset Allocation readjust proportions on the basis of signals (and rules) that indicate the asset will perform better in upcoming period Can utilize options to put a floor on PF value More generally, rules based structured investment methods 1.19 Active Investment Management Objective: Perform better than market or a benchmark chosen as reference Rationalizing Assumption: Financial markets are not efficient with theoretical perfection Markets require time to reflect new information Active management involves strategies to take advantage of temporary inefficiencies Choice of securities for PF: Essential stage for this technique Security selection based on valuation models

6 Active Investment Management Rationalizing Assumption: Financial markets are not efficient with theoretical perfection (continued) These PF have fewer securities (than passive funds) as detailed research is time consuming Can be practiced at asset class level vs. security level (active asset allocation vs. tactical allocation) This method is traditional mutual fund style 1.21 PF Management Organization & the Investment Management (IM) Process Approaches Bottom-Up: Stock Picking; the traditional approach Top-Down: Market and Asset Choices Empirical Research suggests Asset Allocation choices contribute the most to PF Performance Top-Down Approach broken up into 3 phases; each are often separately managed Later, quantitative methods for implementation 1.22 PF Mgmt Organization & the IM Process 3 Phases of the IM Process Strategic Asset Allocation (least amount of time) Contributes to a significant share of PF performance Not just Asset Classes, but also Style and manager selection Tactical Asset Allocation (adjusting asset weightings) Regular, systematic adjustment of PF; capture short-term opportunities; stay close to initial allocation; cf. market timing Stock Picking (most amount of time) The phase where managers specializing in specific asset types ply their trade; valuation models and PF Optimization 1.23 PF Mgmt Organization & the IM Process Performance Analysis Final Stage in the PF Mgmt process Evaluates success of the IM Process Identify contribution of each phase to overall result Provides the means for judging the qualities of a manager Measures the value-add of an active strategy compared with the simple replication of an index or benchmark Performance Analysis Covers all the Techniques implemented to study the results of PF Mgmt Performance Measurement: Quantify Past Results Performance Analysis: Explain Past Results

7 PF Mgmt Organization & the IM Process Performance Analysis Performance Measurement: Quantifies the beginning and ending differential for an evaluation period Performance Attribution: Return due to each phase and the manner within each phase that performance was achieved Performance Evaluation: The skill vs. luck question Contributes to differentiating return from allocation of assets or stock picking, vs. the benchmark 1.25 PF Mgmt Organization & the IM Process Difficulty in Attributing Performance The breakdown of the IM Process into 3 phases requires the comparison with a benchmark Benchmark must be adapted to the PF composition The Benchmark must be constructed w/precision as market indices are rarely representative of the investment strategy Makes real comparison of asset managers difficult The media has its own categories for performance ranking Doesn t account for real risk exposure or portfolio style 1.26 Analysis of the performance of (active) managers funds, compared with the performance of the market, allows (a form of) market efficiency to be tested Managers can add value by determining the gap between the price of a security in the market and its equilibrium value Existence of the gap depends on the time it takes the market to integrate new information and therefore the level/existence of efficiency 1.27 The Market is (Price) Efficient if the prices of assets at any moment reflect all available information Several forms (defined by Fama in 1970) are common. Market Efficiency is: Strong if all information (public and private) is contained in asset prices Following this hypothesis, it is not possible to achieve better performance than the market This is the form that corresponds to the general notion of market efficiency

8 is: Semi-Strong if the prices only reflect public information Weak if today s prices reflect information contained in past prices This form allows for the existence of active investment possibilities for one who has additional/better information There is the possibility to achieve better performance than the market 1.29 A variation on Fama s theme has been found to be: Markets are efficient if the additional profit produced by actively managing a portfolio compensates exactly for the management fees (Grossman & Stiglitz) Corollaries to this are: Investors may either simply observe market prices or they may look for and analyze information The former should practice passive investment The later 1.30 The later can hope to obtain greater profits, but on average the excess profits compensates exactly for the cost of the search This means that the net performance on average is equivalent to that of uninformed investors If everyone adopted a passive strategy, then certain securities would be undervalued Then, one who practiced analysis would be successful The search for information, undertaken to improve performance, plays a role in regulating markets and contributes to their efficiency 1.31 Some results of other studies Fund managers hold securities that beat the market by a quantity that is sufficient to cover their expenses & transaction costs Funds with a high turnover (and high transaction costs) hold securities with a higher (average) return than that of a fund with low turnover This reflects the stock picking skills of managers in very active funds

9 A Model Let r be the risk free rate Assume investors are risk neutral Let R t denote the total return (capital gains and payouts) on an asset in period t to t+1 The efficient market hypothesis (EMH) says E( R I ) (1 r) t t t Where E is the expectation w/r to a given information set I t that is available at time t and includes r t 1.33 A Model Alternatively, the EMH in terms of prices follows from R p p Where p t is the price at t and if there are no payouts E p 1 t 1 It rt pt Or, equivalently, that discounted prices must follow a martingale 1 E( pt 1 It) pt (1 r ) t / t t 1 t 1.34 A Model This comes back to Fama as we specify the (nested) information sets that are used to determine prices For strong-form efficiency, the information set used by the market to determine price contains all information (public & private) that could possibly be relevant to pricing the asset In the case of semi-strong-form efficiency, the set contains publicly available information In weak-form efficiency, only current and past prices are in the information set E p p, p,... 1 r p t 1 t 1 t 2 t t 1.35 A Model An empirical implication of the EMH, often cited as a defining characteristic, is that an efficient price series should move randomly More precisely, in this context, that price changes should be serially uncorrelated, or cov( Rt 1, Rt) 0 t HW for Feb 4 th : show this is so

10 Performance Persistence notwithstanding, Can active portfolio management create added value? Can some of the winners always be the same? Are some managers more skilful than others? If certain managers beat the market regularly, over a statistically significant period, they prove that active management makes sense, Does that not cast doubt over the market efficiency paradigm? 1.37 Performance Persistence A manager who beats the market regularly by taking advantage of arbitrage opportunities from very temporal inefficiencies will not prove that the market is inefficient over a long period The real question is Is investment performance the fruit of real skill or the consequence of luck? Not a question of showing if the markets within which they invest are inefficient 1.38 Performance Persistence In practice, is a manager who has performed well in one year more or less likely to perform well the next? Financial Press Fund Rankings are suggestive Empirical studies are contradictory For study periods of Long Term (3-5 years) and Short Term (1 month or less) there is a trend reversal: Losers become winners and vice versa For Medium Term (6-12 months) winners and losers conserve there characteristics Performance Persistence Empirical studies are contradictory Persistence is more likely in bond funds than stock funds Performance persistence is likely due to the use of momentum strategies rather than stock picking skill Relative, risk-adjusted performance persists While contribution of stock picking to performance in an efficient market is questionable, same cannot be said for asset allocation is important in building performance Empirical studies on performance persistence (efficiency) have enabled performance measurement Poor performers are more likely to persist Managers who perform consistently better do exist 1.39 models to be developed and improved

11 Performance Analysis & AIMR Standard Association for Investment Management and Research (AIMR aimr.org) Performance Presentation Standard (PPS) AIMR-PPS implemented in US Suitable for most developed markets Global Investment Performance Standard (GIPS) A minimal worldwide standard Country Versions of GIPS (AIMR-PPS is one) Investment Firms obtain AIMR-PPS/GIPS certification None include risk-adjusted performance measures 1.41 International Investment Additional Elements Domestic (Here) vs. International (There) vs. Global (Everywhere) Another dimension of diversification and performance Though maybe less-so in an increasingly global economy Consider correlations between economies & markets Though, another dimension of Risk 1.42 International Investment Additional Elements Other dimensions of Risk Exchange risk and political risk Political risk may be more pronounced in developing countries Exchange risk quantifiable Diversify into Intl markets with lower levels of correlation to improve the PF efficient frontier International Investment Additional Elements Return on an Intl PF is the result of the Geographical Factor Manager controls the Currency Factor Cancelled through hedging Managed Separately through a currency performance overlay