scholarly journals A Test for the Number of Factors in an Approximate Factor Model

1993 ◽  
Vol 48 (4) ◽  
pp. 1263-1291 ◽  
Author(s):  
GREGORY CONNOR ◽  
ROBERT A. KORAJCZYK
Keyword(s):  
Factor Model ◽  
Number Of Factors ◽  
Approximate Factor Model

A random-perturbation-based rank estimator of the number of factors

Biometrika ◽  
2020 ◽  
Author(s):  
Xinbing Kong
Keyword(s):  
Limit Theorem ◽  
Factor Model ◽  
Random Perturbation ◽  
Testing Procedure ◽  
Simulation Studies ◽  
Number Of Factors ◽  
Rank Estimator ◽  
Hypothesis Testing Procedure ◽  
Approximate Factor Model ◽  
New Hypothesis

Summary We introduce a random-perturbation-based rank estimator of the number of factors of a large-dimensional approximate factor model. An expansion of the rank estimator demonstrates that the random perturbation reduces the biases due to the persistence of the factor series and the dependence between the factor and error series. A central limit theorem for the rank estimator with convergence rate higher than root $n$ gives a new hypothesis-testing procedure for both one-sided and two-sided alternatives. Simulation studies verify the performance of the test.

Determining the Number of Factors When Population Models Can Be Closely Approximated by Parsimonious Models

2021 ◽  
pp. 001316442199283
Author(s):  
Yan Xia
Keyword(s):  
Factor Model ◽  
Population Level ◽  
Parallel Analysis ◽  
Model Data ◽  
Behavioral Studies ◽  
Number Of Factors ◽  
Parsimonious Models ◽  
Parsimonious Model ◽  
Fit Index ◽  
Tucker Lewis Index

Despite the existence of many methods for determining the number of factors, none outperforms the others under every condition. This study compares traditional parallel analysis (TPA), revised parallel analysis (RPA), Kaiser’s rule, minimum average partial, sequential χ2, and sequential root mean square error of approximation, comparative fit index, and Tucker–Lewis index under a realistic scenario in behavioral studies, where researchers employ a closing–fitting parsimonious model with K factors to approximate a population model, leading to a trivial model-data misfit. Results show that while traditional and RPA both stand out when zero population-level misfits exist, the accuracy of RPA substantially deteriorates when a K-factor model can closely approximate the population. TPA is the least sensitive to trivial misfits and results in the highest accuracy across most simulation conditions. This study suggests the use of TPA for the investigated models. Results also imply that RPA requires further revision to accommodate a degree of model–data misfit that can be tolerated.

scholarly journals Assessing portfolio and asset returns of some financial and non- financial companies on the Ghana stock exchange using a 3-factor model

2021 ◽  
Vol 27 (2) ◽  
pp. 193-202
Author(s):  
C.P. Ogbogbo ◽  
N. Anokye-Turkson
Keyword(s):  
Factor Model ◽  
Stock Exchange ◽  
Asset Returns ◽  
Good Representation ◽  
Financial Firms ◽  
Number Of Factors ◽  
Equity Ratio ◽  
Book To Market Ratio ◽  
Ghana Stock Exchange ◽  
Market Volume

This study on the Ghana Stock Exchange (GSE), investigated, if the overall size of the market, affects the fundamentals of the Fama French 3-Factor model, and to ascertain if the Fama French model can be used effectively to assess portfolio and assets return for companies listed on the Ghana Stock Exchange. In this paper, portfolios of assets of companies on the Ghana Stock Exchange are constructed and analyzed using the Fama-French 3-factor model. The empirical data which consists of assets of 15 companies listed on the GSE, including assets of both financial and non-financial companies for good representation of the Ghana Stock Exchange. We found that the basic principle of the model is not satisfied. This is attributed to a number of factors which include overall size of the market, volume of trade, and high leverage (more debt than equity) associated with financial firms. High debt/equity ratio is linked to high risk. Keywords: Market Capitalization, Book-to-market ratio, Portfolio, Small minus big, High minus low

THE GENERALIZED DYNAMIC FACTOR MODEL: REPRESENTATION THEORY

2001 ◽  
Vol 17 (6) ◽  
pp. 1113-1141 ◽  
Author(s):  
Mario Forni ◽  
Marco Lippi
Keyword(s):  
Latent Variables ◽  
Factor Model ◽  
Companion Paper ◽  
Dynamic Factor ◽  
Dynamic Factor Model ◽  
Data Sets ◽  
Full Characterization ◽  
Approximate Factor Model ◽  
Firm Basis

This paper, along with the companion paper Forni, Hallin, Lippi, and Reichlin (2000, Review of Economics and Statistics 82, 540–554), introduces a new model—the generalized dynamic factor model—for the empirical analysis of financial and macroeconomic data sets characterized by a large number of observations both cross section and over time. This model provides a generalization of the static approximate factor model of Chamberlain (1983, Econometrica 51, 1181–1304) and Chamberlain and Rothschild (1983, Econometrica 51, 1305–1324) by allowing serial correlation within and across individual processes and of the dynamic factor model of Sargent and Sims (1977, in C.A. Sims (ed.), New Methods in Business Cycle Research, pp. 45–109) and Geweke (1977, in D.J. Aigner & A.S. Goldberger (eds.), Latent Variables in Socio-Economic Models, pp. 365–383) by allowing for nonorthogonal idiosyncratic terms. Whereas the companion paper concentrates on identification and estimation, here we give a full characterization of the generalized dynamic factor model in terms of observable spectral density matrices, thus laying a firm basis for empirical implementation of the model. Moreover, the common factors are obtained as limits of linear combinations of dynamic principal components. Thus the paper reconciles two seemingly unrelated statistical constructions.

Polish adaptation of the Geneva Emotional Music Scale: Factor structure and reliability

2020 ◽  
pp. 030573562092747
Author(s):  
Maria Chełkowska-Zacharewicz ◽  
Maciej Janowski
Keyword(s):  
Factor Structure ◽  
Factor Model ◽  
High Reliability ◽  
Factor Solution ◽  
Linguistic Differences ◽  
Number Of Factors ◽  
Confirmatory Factor ◽  
Measurement Tools ◽  
Polish Adaptation ◽  
Being Moved

One of the most popular measurement tools used in music-emotion studies is the Geneva Emotional Music Scale (GEMS). The authors conducted a series of studies on Polish samples to confirm the factor structure and the reliability of the Polish adaptation of the GEMS (GEMS-PL). Study 1 ( n = 262) revealed in the confirmatory factor analysis a good fit for both 9- and 10-factor models, with an indication to a better fit of the 10-factor solution. Additional statistical analyses were performed to explore the differences in the number of factors between the GEMS-PL and the original GEMS. Study 2 ( n = 944) performed in laboratory and Internet settings, confirmed the 10-factor model and revealed that all obtained scales had high reliability. The GEMS-PL scales peacefulness, tenderness, tension, joyful activation, nostalgia, sadness, power, transcendence, and wonder correspond with the original GEMS. The additional scale being moved abstracted from the sadness factor is discussed in the context of possible linguistic differences and growing literature on the state of being moved.

A Simulation of Factor Analytic Reliability Varying Sample Size and Number of Variables

1979 ◽  
Vol 45 (2) ◽  
pp. 471-478 ◽  
Author(s):  
George E. Manners ◽  
Donald H. Brush
Keyword(s):  
Sample Size ◽  
Factor Structure ◽  
Random Sampling ◽  
Factor Model ◽  
Sampling Error ◽  
Factor Models ◽  
Factor Analytic ◽  
Number Of Factors ◽  
Alpha Factor ◽  
Analytic Models

An examination of four factor analytic models employing random sampling experiments is undertaken using a methodology and hypothetical population factor structure first employed by Browne (2). The factor models are each explored under four separate conditions, varying sample size and number of variables. Under these limited conditions, it is argued that there are no practical differences among the factor models considered with respect to sampling error in the absence of a Heywood variable. However, with respect to the ability of each model to capture, early and at convergence, the number of factors in the population, the alpha factor model is shown to have the greatest reliability.

Sign in / Sign up

Export Citation Format

Share Document