CAPM essay 2 - Multifactor models Flashcards

(57 cards)

1
Q

Who created the CAPM

A

Sharpe (1964), Lintner (1965), and Mossin (1966)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is the CAPM model and its aim?

A

Single factor financial model.
Describe the relationship between expected returns and risk.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Why has the CAPM been criticized?

A

Because it considers market risk as the only factor of systematic risk.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Fama and French (2015) findings about the CAPM

A

They find no correlation between average returns and betas in U.S stock returns.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What’s the motivation for multifactor models?

A

CAPMs poor empirical performance and many anomaly results.
Other factors that may influence returns more or alongside systematic risk.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Two main theoretical models when estimating returns are?

A

ICAPM (Intertemporal Capital Asset Pricing Model)
APT (Arbitrage Pricing Theory)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

How is ATP and ICAPM different from CAPM?

A

These models also include other unspecified factors

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Who developed the APT?

A

Ross(1967)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is the Arbitrage Pricing Theory?

A

A multifactor model that can allow for multiple unspecified factors

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What do investors believe under the APT model?

A

That returns are driven by a linear K-factor statistical model.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is K?

A

The number of factors in the model.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

How does the model split unanticipated stock returns?

A

Between K-factors are a residual term (idiosyncratic risk)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

How is the ATP model tested?

A

Time-series regression

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What does this time series regression involved when testing APT?

A

Regressing a stocks returns on the different factors to see how sensitive it is to each one.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

How can a time-series regression be done when the factors are unknown?

A

By regressing returns on chosen factors, even though the theory does not specify what the K factors are

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What do the results of the time-series regression show?

A

How well the factors explain returns based on how strong and significant the relationships are

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

ATP vs CAPM

A
  1. Can be argued ATP is a more general model of CAPM.
  2. ATP doesn’t rely on strong assumptions about investor behaviour.
  3. ATP allows for multiple sources of systematic risk.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Limitations of ATP?

A

1.Doesn’t identify the common K factors.
2.Doesn’t specify the importance of each factor.
3. Isn’t able to show the size or signs of each factors impact.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What are the different approaches when identifying the K-factors according to Ferson (2003).

A

Statistical
Macroeconomics
Portfolio factors

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Statistical factors

A

Come from statistical methods like principal components and don’t always have a clear economic meaning.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Macroeconomic factors

A

Based on things like interest rates, inflation, GDP growth.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Portfolio factors

A

Built using firm features like size or value. such as SMB, and HML.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Who developed the ICAPM?

A

Merton (1973)

24
Q

Why was the ICAPM developed?

A

To extend the CAPM to a multifactor model.

25
Assumptions of the ICAPM?
Investors have multiperiod investment horizons and that they are concerned about changes in future investment opportunity set.
26
Differences in investors: CAPM vs ICAPM.
Investors are not mean-variance optimizers in the ICAPM
27
How is systematic risk captured in ICAPM?
Systematic risk is captured in the ICAPM by beta AND the S state variable.
28
ICAPM critiques
Same as CAPM, Roll Critique(1977)
29
Roll critique (1977)
CAPM is not testable since the market portfolio cannot be observed
30
Other multifactor models
Fama and French 3-factor model (1993) Fama and French 5-factor model (2015) Carhart's 4-factor model (1997)
31
Fama and French 3-factor model (1993)
Three factors are 1.Maret risk 2.Size (SMB) 3.Value (HML)
32
Size, SMB(small minus big)
SMB= Calculated as the return on small-cap stocks minus the return on large cap stocks.
33
Value, HML (High minus Low)
Measures return on high book-to-market (value) stocks minus the return on low book-to-market (growth) stocks
34
HML updated
Asness and Frazzini (2013) updated HML later to reflect more current valuations. Aiming for a timelier version.
35
Carhart's 4-factor model (1997)
Builds on Fama and French 3-factor model (1993). Includes momentum.
36
Momentum, WML (winners minus losers)
Captures the return difference between stocks that performed well in the past and those that performed poorly
37
Pastor and Stambaugh (2013)
Propose a liquidity factor which would show how sensitive a stocks return is to market wide liquidity conditions.
38
Fama and French 5-factor model (2015)
Revisiting the 3 factor model. Now contains RMW and CMA
39
RMW (robust minus weak)
Measures profitability. Firms with strong profitability minus firms with weak profitability
40
CMA (conservative minus aggressive)
Reflects a companies investment. Looks at the return spread between companies that invest conservatively versus aggressively investing.
41
How to test these models?
Also tested with time-series regression
42
What does time-series regression mean?
Regressing stock or portfolio returns on the factor returns to see how exposed they are to each factor
43
The intercept time series regression
Is the alpha, showing if there are any abnormal returns.
44
What does the size of alpha mean?
A small or zero alpha means the model explains returns well.
45
Fama and French 5-factor (2015) flaws
While this model captures average returns better it does still leave out momentum.
46
Why are the ICAPM and the APT considered strong?
They are considered strong financial models because they provide solid theoretical frameworks for explaining asset pricing in real-world markets
47
ICAPM positive summary
Improves on CAPM (multiple time periods and changes in investment opportunity over time).
48
APT positives
Explains returns using several macroeconomic factors. More flexible than CAPM- multiple unspecified factors. Fewer assumptions about investor behaviour
49
How do Fama-French five-factor model (2015) and the Carhart four-factor model (1997) improve on earlier models?
They explain more of the variation in stock returns
50
What are these multifactor models based on?
Observed market patterns, making them highly practical and widely used.
51
What are these models especially good at?
Capturing known anomalies in the CAPM like size, value profitability and momentum.
52
Conclusion - Multifactor model motivation
To conclude while the CAPM laid the foundation for understanding the relationship between risk and return, its limitations have led to the development of more advanced models.
53
Improvements offered by ICAPM and APT?
ICAPM and APT offer strong theoretical improvements by incorporating multiple risk factors and adapting to more realistic investor behaviour
54
Why have empirical models like Fama and French and Carhart become popular?
Become popular in practice due to their ability to explain return patterns and market anomalies more effectively
55
What model is best?
Ultimately, no single model is perfect. Fama French and Carhart still leave unexplained anomalies.
56
How do empirical models work in the real world?
Empirical models often work well in sample but may not always perform in the real world
57
Closing statement- what does the evolution from CAPM to multifactor models reflect?
The evolution from CAPM to multifactor models reflects ongoing efforts to better understand and predict asset prices in increasingly complex financial markets