在財務領域中有許多配適資產報酬率的訂價模型, 但是很少有將數個模型放在
一起比較的相關討論。非巢式的模型定式檢定提供我們比較這許多模型的計量工
具。在許多的資產定價模型中, 我們選擇了較常見的四個模型做為比較的標的, 分
別是Fama-French三因子模型、Ferguson and Shockley 的模型、Liu’s 流動性
二因子模型以及Petkova 的模型。檢定的結果因不同的投組報酬而異。當使用以
size 及B/M 排序而得的二十五投組報酬時, 結果呈現出愈晚被提出來的模型其
解釋報酬率的能力也愈高。當使用十二個產業別的投組報酬時,Liu’s 流動性二因
子模型則較為弱勢。

Many asset pricing models have been proposed to fit the asset returns
in the literature, but few studies discuss the relative performance between
them. Non-nested model specification tests provide the econometric tools to
compare the performance of many models. Of many asset pricing models,
we choose popular Fama-French three factor model, Ferguson and Shockley’s
model, Liu’s liquidity-augmented model and Petkova’s model as competing
models. The results are different with 25 size- and B/M- sorted returns and
12 industry returns. When using 25 portfolio returns, the result has the
pattern that the later proposed model has the better ability to explain the
variation on returns. While the 12 industry returns is used, Liu’s liquidityaugmented
model tends to be the worst model to fit the returns.

References
Ayuda, Maria-´ısabel and Aznar, Antonio (2000), “Power in non-nested models:
a comparative study”, Applied Economics Letters, 7, 483–486.
Chang, Candie Y., Fu, Yuning, and Ng, Lilian K. (1999), “Anomalies, risk,
and japanese stock returns”, .
Davidson, Russel and Mackinnon, James G. (1981), “Several tests for model
specification in the presence of alternative hypotheses”, Econometrica, 49,
781–793.
(1982), “Some non-nested hypothesis tests and the relations among
them”, Review of Economic Studies, 49, 551–565.
(1983), “Testing the specification of multivariate models in the presence
of alternative hypotheses”, Journal of Econometrics, 23, 301–313.
(1990), “Specification tests based on artificial regressions”, Joumal
of the American Statistical Association, 85, 220–227.
(2000), “Bootstrap tests:how many bootstraps?”, Econometric Reviews,
19, 55–68.
Elyasiani, Elyas and Nasseh, Alireza (2000), “Nonnested procedures in econometric
tests of asset pricing theories”, Joumal of Financial Research, 23,
103–128.
Fama, Eugene F. and French, Kenneth R. (1993), “Common risk factors in
the returns on stocks and bonds”, Journal of Financial Economics, 33,
3–56.
Fan, Yanqin and Li, Qi (1994), “Bootstrapping j-type tests for non-nested
regression models”, Econometrics Letters, 48, 107–112.
Ferguson, Michael F. and Shockley, Richard L. (2003), “Equilibrium ”anomalies””,
Journal of Finance, 58, 2549–2580.
Godfrey, L.G. (1998), “Tests of non-nested regression models some results
on small sample behaviour and bootstrap”, Journal of Econometrics, 84,
59–74.
Lewellen, Jonathan, Nagel, Stefan, and Shanken, Jay (2006), “A skeptical
appraisal of asset-pricing tests”, .
Liu, Weimin (2006), “A liquidity-augmented capital asset pricing model”,
Journal of Financial Economics, 82, 631–671.
McAleer, Michael (1983), “Exact tests of a model against nonnested alternatives”,
Biometrika, 70, 285–288.
Park, Teasung (1993), “Equivalence of maximum likelihood estimation and
iterative two-stage estimation for seemingly unrelated regression models”,
Communications in Statistics - Theory and Methods, 22, 2285–2296.
Pesaran, M. Hashem and Weeks, Melvyn (1999), “Non-nested hypothesis
testing: An overview”, .
Petkova, Ralitsa (2006), “Do the fama-french factors proxy for innovations
in predictive variables?”, Journal of Finance, 61, 581–612.
Vuong, Quang H. (1989), “Likelihood ratio tests for model selection and
non-nested hypotheses”, Econometrica, 57, 307–333.
Zellner, Arnold (1962), “An efficient method of estimating seemingly unrelated
regressions and tests for aggregation bias”, American Statistical
Association Joural, 57, 348–368.