Sze Yuen Fung
Macquarie University
Sydney
Australia
INVESTMENT ADVICE
General investment advice to an
investor on how to allocate funds for a long investment horizon of 10-15 years
and a short horizon of 1-2 years?
focusing on bonds and equity.
CONTENT
3) GENERAL SUGGESTIONS FOR LONG-TERM
& SHORT-TERM INVESTORS
4) HORIZON EFFECTS ON ASSET ALLOCATION
4.1) Asset
allocation under the standard CAPM
4.4)
Predictability of stock returns
4.5)
Investment returns under various portfolio combinations
5.1)
Contributions of asset allocation and market timing?
5.2) Costs and
benefits of market timing
5.3) An example
of marketing timing strategies ?Filter rules
5.4)
Effectiveness of filter rules
6) FACTORS AFFECTING ASSET ALLOCATION
6.4) Non-financial (Personal) risk and
benefit
Samuleson (1969) and Merton (1969) both
showed that if returns are independent over time, investors will always choose
the same asset allocation regardless of investment horizon. However, Lo and
MacKinlay(1998), Conrad and Kaul (1998) as well as Jegadeesh and Titman (1993)
showed that stock returns are positively autocorrelated. This makes stocks
become riskier in the short run. Moreover, Fama and French (1988), Poterba and
Summers (1988), De Bondt and Thaler (1985) as well as Barberis (1999) found
that stock returns are mean-reverting in the long run. This makes stocks become
safer in the long run. Pryor and Biller (2002) showed that stocks is safer in
the long run than in the short run and offer a higher average return than bonds
do. Siegel (1998) also suggested that stock market is safer in the long term.
Therefore, this lead to the suggestion that investors with a long horizon
should allocate a higher proportion to stocks. Brinson, Hood and Beebower
(1986) found that about 90% of the quarterly variation in total fund returns
are explained by asset allocation. Brinson, Beebower together with Singer
(1991) do the same study again and also have similar findings. A more updated
study by Ibbotson and Kaplan (2000) also found that asset allocation explains
about 90% of the period-to-period variability of a portfolio return. Market
timing is not an important factor affecting the performance of a fund. Fama and
Blume (1966), Jensen and Benington (1970), and Ball (1978) performed tests on
many filter rules which are market timing strategies using the U.S. stock
market and found that the returns can be negative after deducting transaction
costs. Fama (1970) suggested that this is the result of an efficient market.
Therefore, investors should concentrate on constructing a portfolio which
minimizes taxes and transaction costs.?
The length of an investment horizon affects the asset allocation. In
fact, it is determined by an investment goal. Asset allocation is also
influenced by risk tolerance and non-financial risks and returns of an investor.
These are the factors one must consider before investing.
Rational investors prefer more returns to less returns and less risks to more risks. However, risks and returns are always negatively related. Thus, tradeoffs between risks and returns must be made when constructing a portfolio. Different asset types offer different levels of return and also have different levels of risk. A portfolio can be constructed with an acceptable level of return and risk by combining various asset types. There are many types of financial asset. However, the focus of this essay is only on bonds and stocks. Bonds have a reputation as conservative investments. They are contractual loans made by investors to institutions that, in return for financing, will pay a premium for borrowing, known as a coupon. Additionally, the bond’s face value is returned to the investor at maturity. ?/span>Stocks have traditionally been regarded as risky assets. They are a form of ownership. Generally, there are no guarantees made to investors about returns of the initial investment. In fact, the profitability of the investment depends almost entirely upon rising stock price. They may be attractive because of their high average returns, but these returns represent compensation for risk. One of the reason why holding stocks is riskier than holding bonds is that When a business or firm is terminated or bankrupt, the proceeds from selling its assets are paid to bondholders first and leftovers are then distributed to shareholders. It is possible that there is no leftover after paying to creditors. In recent years, however, it has become commonplace to argue that stocks are actually relatively safe assets for investors who are able to hold for the long term. In other words, the proportion invested in stocks is affected by the length of the investment horizon. In this essay, suggestions about asset allocation for long-term and short-term investors will first be made. This will then be followed by explaining the reasons behind those suggestions through looking at the horizon effects. The importance of market timing and factors affecting asset allocation will also be discussed.
For the most part, investments offering
the greatest growth potential also have the greatest risk. Investors with a
short time horizon should avoid higher risk investments such as stocks or stock
funds, since the growth potential offered by these investments over times can
be offset by short-term volatility. Investors with sufficiently short time
horizon, say one to two years, should concentrate on more stable investments
such as bonds or bond funds. Investors with a long time horizon, say ten to
fifteen years, can generally afford to invest more aggressively in stocks
because short-term volatility will usually be overcome by growth in the long
run. However, the long run can turn out to be extraordinarily long, far longer
than an investor's investment horizon. Therefore, investors should make
themselves clear that holding stocks for a long term is safer but not riskless.
Moreover, there is no one asset allocation mix that is right all the time. For
example, a long-term 15-year horizon will become a short-term 5- year horizon
after 10 years. As the time horizon shifts, asset allocation should shift
accordingly. Therefore, investors should review their asset allocation mix at
regular intervals to ensure meeting their financial goals. Moreover, investors
should not time the market by themselves, however should put more effort in
constructing a portfolio which minimizes taxes and transaction costs.
In section 2, general suggestions have been made that investors with a long time horizon should allocate a higher proportion to stocks and those with a short horizon should allocate a higher proportion to bonds. In this section, the reasons why the length of the investment horizon affects the asset allocation between stocks and bonds will be explained.
The standard CAPM recognises the fact that risk tolerance affects asset allocation. Under the CAPM, every rational investor should solely hold a combination of the riskless asset and the market portfolio, while the composition of the market portfolio is the same for every investor under the assumption of homogeneous expectations. The proportion invested in these two investments is determined by each investor’s risk-tolerance. The more risk-tolerant investors will invest a higher proportion in the market portfolio and the less risk-tolerant investors will invest a higher proportion in the risk-free asset. However, the CAPM does not tell us how investment horizon affects asset allocation. Under the CAPM, asset prices are assumed to follow random walks and are thus unpredictable. Suggestion that stocks are safer in the long run cannot be made since a price drop makes it no more likely that the prices will rise in the future. If returns are independent over time, then the mean and variance of continuously compounded returns rises in proportion to the time horizon. Thus, the ratio of mean to variance is the same at all horizons. Investment horizons affecting asset allocation is called horizon effects. Samuleson (1969) and Merton (1969) showed that if assets are independently and identically distributed, an investor who rebalances his portfolio optimally should choose the same asset allocation, regardless of investment horizon. If market returns are draws from the same random distribution every year, then this suggestion makes no sense. In fact the probability of losing money falls with the horizon, however this is offset by the increasing size of possible losses over long periods. In other words, if the annualised risk of stock market investing is independent of the investment horizon, there is no point in suggesting that holding stocks is safer in the long term.
However, historical evidence suggests that stock returns are related to past returns and are momentum in the short run. The momentum effect implies positive serial correlation of returns. It means that a high return today implies a higher return in the future, vice versa. Another interpretation is that past winners will continue to win and losers will continue to lose. One famous study about the short-term serial correlation of stock returns was conducted by Lo and MacKinlay (1988). They found that there is a positive correlation between past return and future return over weekly periods. It means that a high return this week implies a high return next week on average. Conrad and Kaul (1998) conducted a similar study and also found that a positive serial correlation exists between weekly returns of stock. According to the study by Jegadeesh and Titman (1993), the momentum effect appears primarily over short horizons of about one to twelve months. In other words, they suggested that buying stocks with a good performance in the past one to twelve months is likely to give investors above-average returns over the next one to twelve months. However, Conrad el at (1998) and Jegadeesh el at (1993) suggested that observed tendency was not significant enough to earn any trading profits after accounting for transaction costs. Although shareholders have the chance that the price of their stocks continues to rise in the near future, they also have the risk that the price of their stocks continues to fall in the short run. If they are with a short investment horizon, the worst case is that they have to sell their stocks at the bottom.
Moreover, evidence shows that stock prices are mean-reverting in the long run. Mean reversion implies that negative serial correlation of stock returns. It means that a high return today implies a lower return in the future, vice versa. A most often quoted study about the long-term serial correlation of stock returns was conducted by Fama and French (1988). They used data from 1926-1985 and found that stock returns display negative serial correlation over long horizons. Other studies by Poterba and Summers (1988) as well as by De Bondt and Thaler (1985) also have similar findings. They suggested that the mean reversion effect appears over horizons of three to five years. In other words, they suggested that buying stocks that performed poorly during the past three to five years is likely to give investors above-average returns over the next three to five years. Over a long horizon, stock prices tend to revert towards an average or mean; fluctuations of stock price tend to average out. If bear markets tend to be followed by bull markets in the long run, holding stocks for a long period is safer than holding stocks for a short period. Shareholders with a short investment horizon may have to sell at the bottom rather than wait for the inevitable recovery. On the other hand, shareholders with a long investment horizon have more time to recover from declines in stock prices.
Therefore, in the case of predictable returns, the investor’s investment horizon may no longer be irrelevant. If returns are predictable, mean and variance no longer can be extrapolated in an easy way because of positive or negative serial correlation. In the case of positive serial correlation, the variance increases faster over time than the return does and stocks are less attractive for the long run. In the case of mean reversion, the opposite is true and stocks are more attractive over long horizons. Barberis (1999) also estimated significant mean reversion in stock markets. This suggests that stocks are safer in the long run and that investors with a long investment horizon hold a greater proportion of stocks in their portfolios. Thus, predictability does have an effect on asset allocation. It is important to take into account the estimation risk of predictability. There is some standard error that leads to uncertainty which is harmful to risk-averse investors. Thus, this uncertainty lowers the optimal allocation to stocks compared to the case with fully predictable returns. The longer the horizon, the smaller the proportion of stocks an investor will choose to hold. In Figure 1, Barberis (1999) showed that for an investment horizon of ten years, the optimal allocation to stocks with uncertain predictability is only half as much as in the case of full certainty. But there is still enough predictability in returns to make investors allocate substantially more to stocks. However, the difference is not as large compared to predictability and full certainty. A long-horizon investor who ignores estimation risk may over-allocate to stocks by a sizeable amount.
Figure 1
The effect of estimation risk and the investment horizon on the optimal allocation to stocks (Source : Barberis 2000)
An easy understanding study by Pryor and Biller (2002) also showed that the longer the horizon, the higher the proportion of stocks an investor should choose to hold. Historical investment returns of various portfolio combinations from 1950 to 2001 were computed using data from the 2002 Yearbook of Ibbotson Associates. Firstly, they looked at the results from buying the assets on January 1, 1950, and holding for twelve months, then they "rolled" to the next month to see what happened if the assets had been purchased on February 1 and held for twelve months. Then they moved to March 1 and did the same thing. And so on. Continuing in this way, they computed the results for all of the 613 twelve-month holding periods from 1950 through 2001. This is in contrast to just 52 periods when only calendar years are considered. Using this more exhaustive process provides a more accurate picture of the degree of volatility and level of returns that can be expected from different blends of stocks and bonds. The whole figure is shown on Figure 2. There are 48 scenarios and each represents a different combination of a holding period and a portfolio mix. Scenario 1 and 6 show that the average annualised return of stocks and bonds over 613 12-month holding periods are 13.8% and 6.5% respectively. The average annualised return of stocks over the best and worst 12-month period are 61% and -38.9% respectively for a range of 99.9%. On the other hand the average annualised return of bonds over the best and worst 12-month period are 32.7% and ?.6% respectively for a range of 38.3%. The average annualised return is higher for stocks, however the range is smaller for bonds. An extremely unlucky investor will have a loss of 38.9% if only holding stocks, however she/he will only lose 5.6% if only holding bonds. In other words, only holding stocks will lose 33.3% more than only holding bonds over the worst 12-month period. Therefore, it is obvious that over the 12-month holding period, holding bonds is relatively safer compared with holding stocks. The picture is very different for long holding periods of 15 years or more. The average annualised return of stocks over the best and worst 15-year period are 19.6% and 4.1% respectively for a range of 15.5%. On the other hand the average annualised return of bonds over the best and worst 15-year period are 11.4% and 2.4% respectively for a range of 9%. An extremely unlucky investor will have a return of 4.1% if only holding stocks, however she/he will only have a return of 2.4% if only holding bonds. In other words, only holding stocks will have a annualised return 1.7% more than only holding bonds over the worst 15-year period. Therefore, for a 15-year holding period, holding bonds seems to be as safe as holding stocks. In addition, holding stocks provide an average annualised return? ( 11.5-7 )% = 4.5% higher than holding bonds. For a holding periods more than 15 years, holding stocks provide a higher return over the best and worst period compared with holding bonds. Thus, with a long holding period of 15 years or more, holding stocks not only provide investors a higher average annualised return but also with a lower risk compared with holding bonds.
Historical investment returns of various
portfolio combinations from 1995 to 2001. (Source : Pryor and Biller 2002)
A study by Siegel (1998) supports this view with a thorough analysis of stock market data, showing that stocks returned about 7% above inflation for the past two hundred years on average. Also, for twenty-year time frames, stocks outperformed bonds over 90% of the time. Siegal (1998) also argues that stocks are actually riskless for long-term investors and should be priced to return no more than Treasury bills. This is a dangerous overstatement, as reduced risk at long horizons does not make stocks a "sure thing". However, the data above also show that the longer is the holding period, the safer is holding stocks compared with holding bonds. The explanation is that the bonds mentioned above are not inflation-indexed government bonds. Bondholders are therefore still facing inflation risk, credit risk and prepayment risk.?Therefore, over a long horizon holding stocks can be safer than holding bonds except inflation-indexed government bonds. Moreover, investors should always remember that the riskless assets for long-term investors are inflation-indexed government bonds, not stocks.
Suggestion that stocks are safer in the
long run than in the short run and investors with a longer horizon should
invest a higher proportion to stocks is based on the evidence that stock prices
are momentum in the short run but are mean-reverting in the long run. It means
that a low return today implies a lower return in the near future but a higher
return in the further future, vice versa. In the short run, there is a
probability that the stock price continues to rise but also a probability that
the stock price continue to fall. It means that there is a probability that
shareholders obtaining huge gains by selling at a top price but also a risk
that they suffering huge losses by selling at a bottom price in the short run.
This risk can be eliminated if they have a long investment horizon because
stock prices tend to revert to a mean. They can wait for a price recovery
instead of sell at the bottom. That is the reason why stock is suggested to be
safer in the long run rather than in the short run. Moreover, the average
annualised return of stocks is higher than that of bonds over both short
horizons and long horizons. That is the reason why investors with a long
horizon are suggested to hold a higher proportion of stocks in their
portfolios.
Investors are said to be timing the market if they attempt to predict future market directions, usually by examining recent price and volume data or economic data, and investing based on those predictions. However, costs must be incurred when collecting those data. In this section, whether it is worth adopting a market timing strategy will be explained.
Many researchers have studied funds to
determine what accounts for variations in returns. A common conclusion of many
of these studies is that a large percentage of the variability is attributable
to asset allocation. The most frequently cited study was by Brinson, Hood and
Beebower (1986). This study looked at the returns of 91 large pension funds in
U.S. over the 1974-1983 period and found that investment policy (asset
allocation) dominated investment strategy (market timing and stock selection),
explaining on average 93.6% of the quarterly variation in total fund returns.
In particular funds asset allocation explained no less than 75.5% and up to
98.6% of the total returns variation. Attempts to actively manage the portfolio
actually cost the average fund 1.1% per year (0.66% due to market timing and
0.36% due to stock selection). In 1991, Brinson, Beebower together with Singer
published a follow-up study looking at the returns of 84 large pension funds in
U.S. over the 1977-1987 period and found that on average asset allocation
explained 91.5% of the variation in quarterly returns. Ibbotson and Kaplan (2000) replicated
Brinson’s results with two new sets of data. They looked at 10 years of monthly
returns for 94 balanced (i.e., stock and bond) mutual funds and 10 years of
quarterly returns for 58 pension funds. They confirmed Brinson’s finding that
asset allocation explains about 90% of the period-to-period variability of a
portfolio return.
Individual investors are not suggested to time the market by themselves. Even the smallest of these pension funds represented a very large investment pool. Therefore, assumption can be made that they commanded the very best talent available. Each was the valued client of one or more of the largest and most prestigious investment managers in the world. As such, they automatically received the best research and information. In other words, they certainly had the resources available to "beat the market." However, the study of Brinson et al. (1986) concluded that on average, attempts to actively manage the portfolios actually cost the average fund 1.10 % per year when compared to just buying and holding the appropriate indexes. This is not to say that active management (market timing and stock selection) is useless. Investors who have the ability to select superior managers before committing funds can earn above-average returns. Individual investors are likely to be less able to beat the market compared with the managers of those large funds and therefore are not suggested to time the market by themselves. In other words, the decision to invest in asset classes (stocks, bonds, etc.) was more important than the market timing.
Filter rules are a market timing strategy. Many studies suggested that filter rules do not generate better returns than a buy-and-hold strategy. Filter rules show investors when they should be long in a security and when they should sell it short. As long as no new information enters the market, the price of a stock fluctuates randomly around the “fair? price.?“Professional investors?will enter and buy or sell the stock if the actual price deviates too much from the “fair?price. Their actions will keep the security price within the two barriers. However, if new information enters the market, then a new equilibrium will be determined. If the news is very favourable, then the price should go up to a new equilibrium. Investors buy the stock as it breaks through the upper barrier will benefit from the price increase to the new equilibrium level. Other the other hand, if the news is unfavourable, then the price will drop to a new equilibrium. Investors can avoid much of the price drop by selling the stock as it breaks through the lower barrier. They can also benefit from the price drop by selling the stock short. The filter rule is usually expressed in the following way : Buy the stock when it increases by X% from the previous low and hold it until it decreases by Y% from the subsequent high. At this point, sell the stock short or hold cash.
Fama and Blume (1966), Jensen and Benington (1970), and Ball (1978) performed tests on many filter rules using the U.S. stock market. All of them concluded that filter rules do not generate superior returns to the buy-and-hold strategy. Moreover, if transaction costs were taken into consideration, the returns could even be negative. These results are consistent with the efficient market hypothesis. This hypothesis implies that technical analysis is without merit. Fama (1970) stated that in an efficient market, all available information including the historical prices and past trading volume is reflected by the current price level. As investors compete to exploit their common knowledge of a stock’s price history, they necessarily drive stock prices to levels where expected returns are exactly commensurate with risk. At those levels no investors can expect abnormal returns.
Investors who want to time the market must incur a cost in doing research and collecting information. The cost tends to be huge but in return the benefit is small. Therefore, investors should not time the market by themselves. Investors who have the ability to select superior managers before committing funds can earn above-average returns. However, there is no guidance showing how to choose a superior manager. Moreover, there is no incentive for fund managers to a market timing strategy since performance measurement is short-term. Their careers may end abruptly when the target is missed once. Thus, investors should put more effort to construct a portfolio which minimizes taxes and transaction costs instead of to time the market.
In this section, factors affecting asset allocation will be discussed without using too much jargons of finance. Examples will be used to make ideas more understandable.
Different investment goals lead to different
investment portfolios. Investment goals are goals that people want to achieve
by using their investment proceed. Saving for retirement, buying a car, buying
a house, and paying for a child’s education are some examples of investment goals.
Each goal will defined the amount and liquidity (ability to turn investments
into cash) of money needed as well as the number of years (investment horizon)
available for investment to grow. As goals are different among people,
different people will have different investment portfolios. For example, the
investment portfolios of retirees, parents of a newborn baby, and newlyweds who
are saving for their first home, will differ markedly.?Retirees often interested in regular and
stable income to cover immediate needs, therefore they tend to consider less
risky investment like bonds. Parents of a newborn baby who are seeking to build
a college fund may want a portfolio predominately composed of growth-oriented
stock funds since their investment horizon is 18 years. For medium-term goals,
such as saving for buying a home, a moderately aggressive approach should be
used and invest in stocks with moderate risk and potential for growth.
Moreover, an individual investor can have more than one goal simultaneously.
Setting up separate investment plans can help accomplishing each goal. To
conclude, clearly defining investment goals is an important step in the asset
allocation process.
Risk tolerance is an important factor in deciding weights of assets included in a portfolio. It is a measure of willingness to trade off volatility of returns (investment risk) for higher average returns. This determines the allocation between risky and risk-free assets. A guideline, which can be used by investors to determined whether she/he is more or less risk tolerant than the average investor, may be the proportion of the overall market that is 60% stocks and 40% bonds. Aggressive (More risk-tolerant) investors are likely willing to accept risk in exchange for higher expected returns and tend to invest in growth stocks. Therefore, they may have a portfolio of 80% stocks and 20% bonds. On the other hand, conservative (less risk-tolerant) investors are less willing to accept risk, even for higher expected returns. Capital preservation is likely to be a top priority for conservative investors. Therefore, they tend to favour conservative investments such as cooperate bonds and government bonds. Thus, they may have a portfolio of 30% stocks and 70% bonds. That is the reason why investors with different levels of risk tolerance will have different portfolios. According to data published by Centre for Research in Security Prices at the University of Chicago, stocks have averaged yearly returns of about 12% over the last 50 years. If investors wish to have annual returns of 15%, they would have to invest in aggressive growth stocks that have greater risk in hopes of achieving their desired return. Moreover, risk tolerance may vary from different investors due to different personality and often changes when going through stages in the life cycle. For example, young adults who have potential for a growing income may be comfortable taking higher risk. Since if they make a financial mistake, there is time to make a correction. Those nearing retirement may feel uncomfortable with risk since time for financial recovery is short. In fact, whether investors are more risk tolerant or less risk tolerant, some risks are suggested to be taken in order to ensure the purchasing power not be eroded by inflation.
Investors with long time horizons should choose investments of higher risk but also offering higher average/expected returns. Those with short horizons should take the opposite approach. Investment/Time horizon is the length of time between when money is invested and when investment proceeds are needed. Time horizon ends when investors plan to liquidate an entire portfolio to meet a goal. As discussed earlier, an investment horizon is determined by an investment goal. For example, young investors who want to build up funds for retirement have long time horizons, and therefore can choose investments that exhibit wide price swings such as growth-oriented stocks because there is sufficient time available for fluctuations to average out. However, a long time horizon will become a short time horizon when they approach their retirement age. At that time, they may wish to transfer some of their assets into less risky investments like government bonds that lock in gains and provide a guaranteed income stream in order to cover after-retirement daily expenses. Therefore, it is important to regularly rebalance asset weights of a portfolio to achieve planned investment goals.
Not only financial risks and returns but also other personal risks and returns should be taken into account when constructing a portfolio. Investors should consider whether they wants to trade off some return against more or less of other priced risks. For example, an investor might want to avoid the risk that her/his portfolio performs poorly at the same time that his industry suffers from a downturn. Therefore, it makes sense for her/him to avoid stocks that are sensitive to recessions. Similarly, investors might want to avoid bonds that will all do poorly in a credit crunch even though they might offer a high yield. The typical investor should hold a stock position that is short his company in order to protect herself/himself against the risks of her/his occupation. However, many employees hold long positions for incentive reasons and in order to have influence on management decisions in their own company. They are trading off non-financial risks for non-financial returns. Therefore, non-financial risks and non-financial returns should also be taken into account in the portfolio construction.
An investment goal determines the length of the investment horizon. Both of them are factors affecting asset allocation. Besides, factors such as non-financial risks and returns, and risk tolerance also can influence a portfolio construction. The serial correlation of stock returns makes the investment horizon becoming an important factor affecting the relative proportion of stocks and bonds included in a portfolio. Evidence shows that stock returns are positively correlated in the short run. This makes investing in stocks becoming riskier in the short run as a stock market investor is exposed to the risk of a bear market. In addition, evidence shows that stock returns are mean-reverting in the long run. This implies that stock market investing is comparatively safer at long horizons as the loss caused by a bear market will likely be compensated, at least in part, by a subsequent bull market. Studies show that the average annualised return of stocks is higher than that of bonds over both the short and long holding periods and holding stocks becomes as safe as or safer compared with holding bonds when the investment horizon increases. Therefore, investors with a long investment horizon should allocate a higher proportion to stocks and those with a short horizon should invest a higher proportion in bonds. Investors may want to adopt a market timing strategy to make a profit. However, studies show that it seems not to be profitable after deducting the transaction costs. They, therefore, suggest that investors should put more efforts in constructing a portfolio which minimizes taxes and transaction costs.
Ball, R. (1978), “Filter Rules : Interpretation of market efficiency, Expermental problem and Australia experience? Acounting Education, 1-7.
Barberis, N. (1999), “Investing for long run when returns are predictable? Journal of Finance, forthcoming.
Brinson, G., Hood, L. and Beebower, G. (1986), “Determinants of protfolio performance? Financial Analyst Journal, 39-48.
Brinson, G., Singer, B. and Beebower, G. (1991), “Determinants of portfolio performance 2 : An Update? Financial Analyst Journal, 40-48.
Conrad, Jennifer, and Kaul, Gautam (1988), “Time-variation in expected returns? Journal of business, 61, No.4, pp409-425.
De Bondt, Werner, and Thaler, Richard (1985), “Does the stock market overreact?? Journal of Finance, 40, pp793-808.
Fama, Eugene, and Blume, Marshall (1966), “Filter rules and stock market trading? Journal of Business, 226-241.
Fama, Eugene (1970), “Efficient Capital Markets : A review of theory and empirical work? Journal of Finance, 25, 383-417
Fama, Eugune, and French, Kenneth R. (1988), “Permanent and temporary components of stock prices? Journal of political Economy, 96, pp246-273.
Ibbotson, R.G. and Kaplan, P.D. (2000), “Does asset allocation policy explain 40, 90, or 100 percent of performance?? Association for Investment Management and Research.
Jegadeesh N. and Titman S.(1993), “Returns to buying winners and selling losers : Implication for stock market efficiency? Journal of Finance, 48, pp65-91.
Jensen, M.C. and Benington, G.A. (1970), “Random walks and technical theories : Some additional evdience? Journal of Finance, 469-482.
Lo, Andrew W., and MacKinlay, A. Craig. (1988), “Stock market prices do not follow random walks : Evidence from a simple specification test? Review of Financial Studies, 1, No.1, pp41-46.
Merton, R. C. (1969), “Lifetime portfolio selection under uncertainty : The continous time case? Review of Economics and Statistics, 51, No. 3, pp247-257.
Poterba, James M., and Summers, Lawrence H. (1988), “Mean reversion in stock prices : Evidence and implications? Journal of Financial Economics, 22, No.1, pp27-59.
Samuelson, P. A. (1969), “Lifetime portfolio selection by dynamic stochastic programming? Review of Economics and Statistics, 51, No.5, pp239-246.
Pryor, Austin, and
Biller, Mark (2002), ?/span>Stocks
vs. bonds: Picking the right mix for your situation? online article of Sonud
Mind Investing company.
http://www.soundmindinvesting.com/vsection/v_newsletter/0211/feature.htm
Siegel, Jeremy (1998), “Stocks for the long Run? 2nd edition, McGraw-Hill.