The Fama And French Three Factor Model Finance Essay

In the portfolio direction field, Eugene Fama and Kenneth French developed the Fama-French three factor theoretical account to depict market behaviour. The FF three-factor theoretical account for the expected return on portion is similar to the CAPM but with two excess factors. Fama and French ( 1992 ) find that asset-pricing theoretical account of Sharpe ( 1964 ) , Lintner ( 1965 ) , and Black ( 1972 ) has some drawbacks, empirical contradictions, refering mean returns. Their research fails to back up the cardinal anticipation of the CAPM theoretical account, that mean stock returns are positively related to market i?? get downing from 1963 boulder clay mid-1990s. Furthermore, variables like size ( market equity ) , purchase, book-to-market equity, earnings-price ratios ( E/P ) can be regarded as assorted ways to scale stock monetary values. They can be considered as different ways of pull outing information from stock monetary values about the cross-section of expected stock returns ( Ball ( 1978 ) ; Keim ( 1988 ) ) . Fama and French ( 1992 ) besides indicate that the combination of size and book-to-market equity seems to absorb the functions of purchase and E/P in mean stock returns, as they seem to be excess for explicating mean returns. All in all Fama and French noticed that two categories of stocks have tended to make better than the market as a whole: ( I ) little caps and ( two ) stocks with a high book-to-market ratio. They so added two factors to CAPM to reflect a portfolio ‘s exposure to these two categories:

Hire a custom writer who has experience.
It's time for you to submit amazing papers!

order now

Rit A – A Rft A = i??im +A i??M, T ( Rmt – Rft ) A + i??SMB, T ( SMBt ) + i??HML, T ( HMLt ) + i??it

Therefore, Ri is the portfolio ‘s rate of return, Rf is the riskless return rate, and Rm is the return of the whole stock market. The “ three factor ” I?M is correspondent to the classical I? but non equal to it, since there are now two extra factors to make some of the work. SMB stands for “ little ( market capitalisation ) subtraction large ” and HML for “ high ( book-to-price ratio ) minus low ” ; they measure the historic extra returns of little caps over large caps and of value stocks over growing stocks. These factors are calculated with combinations of portfolios composed by graded stocks and are available historical market informations. Historical values may be accessed on Kenneth French ‘s web page.[ 1 ]

Fama and French attempted to better step market returns and, A through research, found that houses with hapless chances, defined by low stock monetary values and high ratios of book-to-market equity, have higher expected stock returns than houses with strong chances. By including these two extra factors, A the modelA adjusts for theA outperformance inclination, which is thought to do it a better tool for measuring director public presentation.

There is a batch of argument about whether the outperformance inclination is due to market efficiency or market inefficiency. On the efficiency side of the argument, the outperformance is generallyA explained by theA surplus hazard that high book-to- market equity and little cap stocksA face as a consequence of their higher cost of capital and greaterA concern hazard. On the inefficiency side, market participants mispricing the value of these companies, which provides the extra return in the long tally as the value adjusts, explicate the outperformance.[ 2 ]

The Chen-Zhang Three-factor Model

Chen and Zhang ( 2009 ) have proposed a new theoretical account that explains a figure of confounding facets about stock returns. Over the past 20 old ages it has become clear that Fama-French theoretical account is non able to explicate many cross-sectional forms. Cheng and Zhang ( 2009 ) actuate a new three-factor theoretical account from q-theory. In the new theoretical account the expected returns on a portfolio in surplus of riskless rate is described by the sensitiveness of its return to three factors: the market factor, a low-minus-high investing factor, and a high-minus-low returns on assets factor:

Ri, t – Releasing factor, t = i??i, T +A i??M, T ( RM, t – Rft ) A + i??INV, T ( INVt ) + i??RAO, T ( RAOt ) + i??it

The public presentation of the theoretical account is outstanding. The new three-factor theoretical account outperforms traditional plus pricing theoretical accounts in explicating anomalousnesss associated with short-run anterior returns, fiscal hurt, net stock issues, plus growing, net incomes surprises, and rating ratios. Chen and Zhang do non construe the investing and ROA factors as hazard factors. Furthermore factors are constructed on economic basicss that are less likely to be affected by mispricing, at least straight. The grounds of Chen and Zhang suggests that low-investment stocks and high-ROA stocks have high mean returns whether or non they have similar return forms of other low-investment and high-ROA stocks.

Descriptive Statisticss


Over the 16 old ages, the highest average return is for portfolio 1 ( 0.010316 ) , while the lowest average return is for portfolio 10 ( 0.005699 ) . Equally far as average are concerned, the highest comes to portfolio 5 ( 0.01625 ) and the lowest comes for portfolio 10 ( 0.00795 ) . The standard divergence of all 10 deciles portfolio reveals that portfolio10 is the least volatile ( 0.041051 ) with portfolio2 being the least volatile ( 0.06326 ) . The lowest minimal returns are for portfolio2 ( -0.2324 ) and the highest minimal returns are for portfolio10 ( -0.1486 ) , while the highest maximal returns are for portfolio 1 ( 0.2916 ) and the lowest maximal returns are for portfolio 6 ( 0.1097 ) .

Talking about the lopsidedness, it is observed that portfolio 1 has positive lopsidedness ( 0.242876 ) , which means that return distribution of that stock has a higher chance of gaining positive returns and that additions are likely to be greater than anticipated by the normal distribution. On the contrary, portfolios 6 every bit good as portfolio 10 have negative lopsidedness, which indicates the higher chance of gaining negative returns. The positive extra kurtosis are spotted in all 10 portfolios viz. the distribution is peaked or leptokurtic relation to the normal which is 3. Therefore, it means that the portfolios can hold more frequent big positive or negative returns.


We imported the informations in EVIEWS after forming them. We so ran a arrested development for each portfolio utilizing the ordinary least squares ( OLS ) method with the dependant variable being the extra returns of the portion over the hazard free rate and the independent variables being the factors that each theoretical account specifies as lending to those returns. We so collected the adjusted coefficients of finding ( ) for every arrested development and constructed graphs demoing the adjusted R2 on the y-axis and the market capitalisation of the portfolio on the x-axis. Graphs comparing the CAPM with each other theoretical account and one embracing them all together were constructed.


The intent of this exercising was to utilize the R2 of the arrested developments to find the power of each theoretical account to construe the variableness in the extra returns following the illustration of Roll. EVIEWS calculates the adjusted R2 as:

The adjusted R2 is a better step of goodness of tantrum than the simple R2 because the latter does non diminish when the figure of independent variables or regressors increases. This would make some uncertainty as some of the theoretical accounts that we use integrated more factors than others. An utmost instance would be to hold a R2=1 if the regressors are every bit many as the sample observations and it has been observed that it can be every bit high as 0.9 for clip series. As we can see from the expression above the R2-adjusted can diminish when we add regressors to the equation and could even turn out to be negative for some theoretical accounts with limited capablenesss. However as Roll mentions the step can non be considered as a conclusive trial as we can non set up whether the factors used are permeant and related to the hazard wagess. ( Roll, 1988 ) . For this ground it is considered to be a “ soft ” step. Furthermore there is no distribution for which will enable us to analyze its behaviour. Finally the can be unnaturally high when there is a tendency in the information. Obviously this can non be a job as our variables are non in degrees but in differences.

The Adjusted R-square Distribution in Four Models

Distribution of Adjusted R-square

It is obvious merely by eyeballing the informations that all of our theoretical accounts have an upward tendency of adjusted R-square, that is additions when the portfolio size additions. This is consistent with Roll`s consequences ( Roll, 1988 ) . High capitalization portfolios can be considered as a aggregation of smaller 1s. These portfolios are diversified and hence they will hold higher R2. In the spirit of Roll we ran a arrested development between R2 and size for all the theoretical accounts to prove for statistical significance

CAPM versus Chan and Zhang

. The Compared with CAPM, Chen and Zhang theoretical account consequences in a larger adjusted R-square in the first six portfolios, with a diminishing difference from p1 to p6. It is so followed by meeting, eventually making about 93 % for both theoretical accounts. However, Chen and Zhang is superior to CAPM in this arrested development, for merely have one smaller adjusted R-square ( p7 ) than CAPM.

CAPM versus Fama French

Compared with CAPM, Fama French shows strong capacity to explicate extra return, whose R-square supports larger for each of the portfolios. As we see, the lowest adjusted R-square is above 60 % and the highest to be 96 % , keeps in a high degree all the clip. However, in common with Chen and Zhang, this dominant place, for FF is threatened by the increasing size of portfolio, which means, with the rise in size, the spread between two theoretical accounts is worsening.


It is amazing that CRR has a instead weak adjusted R-square, which somewhat fluctuates around nothing. There ‘s even a negative figure in p2 bespeaking bad arrested development consequences. Such a low R-square agencies those CRR factors has small capacity to explicate extra return, There is besides no form in graph compared with CAPM ‘s upward tendency which can non uncover any relationship between size and R-square.

Comparison and Conclusion

From the graph above we can see that there is a form or instead, a positive correlativity between portfolio size and R-square, with larger portfolios holding higher R-squares with the lone exclusion of the APT theoretical account where the R-square is non correlated with the size of the portfolio. For illustration the highest R-square in the APT theoretical account is for the decile 6. This behaviour may propose as proposed by Roll ( 1988 ) that large market capitalisation stock portfolios are less sensitive to the factors that attempt to explicate the extra returns in the APT theoretical account.

CRR APT works worst, with highly low adjusted R-square

CRR is ill worse than other three, for its R-square ne’er goes beyond 3 % through 10 portfolios. This indicates that there is merely a minor part of mean extra return which can be explained by CRR factors. Therefore, whether the arrested development is dependable should name for farther testing.

For good theoretical accounts, adjusted R-square additions with the growing in size

As we see, all of the three good theoretical accounts have an uptrend in gait with the addition of portfolio size. There seems to be a statistically important connexion between adjusted R-square and size, which indicates that these theoretical accounts works better for larger size portfolios. Diversification may be a reansonable account for this, because larger houses by and large have many divisions and frequently operate in more than one industry or market. Diversified portfolios should hold high R-squares, for some hazards have been diversified.[ 3 ]

Fama French, Chen & A ; Zhang does good for big size portfolio every bit good as little size

Fama French has a really high get downing point from the smallest size, traveling up aggressively followed by smooth addition. Chen & A ; Zhang ‘s line appears somewhat different, for it moves up in a more stable manner. Overall, both of them have a great grade of explanatory power for big size every bit good as little size portfolio. Contrasted with them, CAPM is much weaker in explicating little size portfolio but works excellent for big size.

As can be seen the Fama & A ; Gallic theoretical account has the most explanatory power of all the theoretical accounts with the highest mean adjusted R-Square. The average adjusted R-Squares were, severally, 0.790 for Chan & A ; Zhang, 0.757 for CAPM, 0.851 for Fama & A ; French and 0.010 for APT. All the 10 size-sorted decile portfolios show a higher R-Square for the FF theoretical account. The APT theoretical account has a really low explanatory power for all the portfolios with even a negative adjusted R-Square for the 2nd decile portfolio. A negative adjusted R-square is given because ( account ) .

Economic ground that could impact the public presentation of the peculiar theoretical accounts in explicating the stock surplus returns have to be explained taking into consideration the period of clip that we are analysing. Our information goes from 1990 to 2006, a period where the US stock markets and many other markets in the universe experienced an extraordinary public presentation due to low involvement rates, low and steady rising prices, the roar in international trade due to globalisation, the engineering roar, and a general economic growing.

From 1980 both, the short term and long term involvement rates have been diminishing whereas industrial production and corporate net incomes have been increasing every bit good as ingestion but maintaining rising prices stable due to an increasing outsource production in emerging markets. This has reduced the cost of goods sold, cut paysheet disbursals and stabilised monetary values in the US market. Besides the monetary value of trade goods remained instead low during this period with for illustration, oil trading at an one-year mean monetary value per barrel of USD 23.19 in 1990, USD 27.39 in 2000 and USD 58.30 in 2006 ( hypertext transfer protocol: // ) , truly far off from the historic high of USD147.30 in July, 2008.

With all this said we think that the theoretical accounts such as Fama & A ; Gallic and Chan & A ; Zhang did better in pricing theses assets because they used explanatory variables that focus more on company specific features such as the ROA factor, the investing factor ( INV ) , little ( market capitalisation ) subtraction large and high ( book-to-price ratio ) minus low and non in macroeconomic factors such as rising prices, return on long term authorities bonds, Treasury Bill rates etc as was the instance with the APT theoretical account. We gave more importance to the factor used by the first two theoretical accounts mentioned because given the economic stableness of the period of survey, macroeconomic factors could be more easy predicted and reflected in the market doing differences in US stock returns being better explained by more specific company related factors


Richard Roll ( 1998 ) , R2, The Journal of Finance, Vol. XLIII, No.2


I'm Heather

Would you like to get such a paper? How about receiving a customized one?

Check it out