Financial Econometrics

CAPM


FAMA-FRENCH MODEL

Problem: CAPM alone doesn’t explain all asset returns.

Solution: Fama and French propose two additional factors:

  1. SMB (Small Minus Big): Return difference between small- and large-cap stocks.

  2. HML (High Minus Low): Return difference between high and low book-to-market stocks.

Model equation (for each asset i at time t):

  • Fama-French 3-factor model is better than CAPM, based on R².

  • However, two important limitations:

    1. Instability of parameters over time.

    2. Non-zero alphas imply incomplete explanation of returns—even with 3 factors.

Risk Analysis


BOOTSTRAP

Problem: Asymptotic results (like normal distribution of estimators) are only reliable with large samples

With small samples, these approximations can be misleading → confidence intervals (CIs) may be too narrow or too wide

Solution: Use bootstrap if dataset = small to simulate a large number of samples & better approximate the sampling distribution


Bootstrap CI Procedure

  1. Estimate the parameter θ from original sample

  2. Create B (large) random samples (with replacement) of the same size (T) from the original data with replacement

  3. Compute θᵇ (estimate) for each sample.

  4. Use the quantiles of the empirical distribution of θᵇ − θ to form a confidence interval

Key Point:

  • Even if we don’t know the exact distribution of θᵇ, we can estimate percentiles (quantiles) from the bootstrap distribution.

  1. In addition, we can argue that we can estimate the distribution of

    This makes sense if we assume that 𝜃_hat arises from sampling from a distribution with parameter 𝜃 & 𝜃^b_hat from 𝜃_hat

→ Bootstrap procedure only works with independent data

If a dependency → Block bootstrap

Block Bootstrap solves dependency issue:

  • Instead of resampling single observations, resample blocks of consecutive data points.

Divide data into k blocks of size l (where T=k×lT = k × l ).

Rule of thumb: l=T(13)l=T^{\left(\frac13\right)}

Preserves local dependency.

Still consistent, but slower convergence than regular bootstrap.

BUT

  • Block bootstrap intervals are wider → means usually not as accurate as the traditional bootstrap

  • Block bootstrap allows to deal with a degree of dependency between observations not possible with the traditional bootstrap


TESTING STATIONARITY

2 types of stationarity

  • Strong (strict) stationarity:

    Joint distribution of observations does not change over time. That is, when the distribution (Yₜ₁, ..., Yₜₙ) has the same distribution as (Yₜ₁₊ₕ, ..., Yₜₙ₊ₕ) for any h

  • Weak stationarity: The series has:

    1. Constant mean

    2. Constant variance

    3. Covariance independent of t

→ In econometrics, we usually test weak stationarity


Rule of Thumb to detect (non-)stationarity

  • Visual check: Plot the time series:

    • If path oscillates around the mean often → likely stationary

    • If there's a trend or the mean shifts → likely non-stationary


Non-stationarity test - Dickey-Fuller

H0:β=0H_0:\beta=0 → unit root = non-stationary

H1:β<0H_1:\beta<0 → stationary

! NOT use standard t-test, because under H0H_0 the process is not stationary !

→ Critical values have been tabulated (Fuller 1976)


Testing Procedure

  • Tests with H₀ = non-stationary:

    Dickey-Fuller ; Augmented Dickey-Fuller (ADF) ; Schmidt-Phillips ; Elliott-Rothenberg-Stock

    • RH0 (p-value < α\alpha ) : reject non-stationarity → stationary series

    • NRH0 (p-value >α\alpha ): non-stationary fails → non-stationary series

  • Tests with H₀ = stationary:

    KPSS test (Kwiatkowski et al.)

    • RH0 (p-value < α\alpha ) : YtY_{t} is not stationary → non-stationary

    • NRH0 (p-value >α\alpha ): cannot say YtY_{t} is stationary → stationary


Models that require stationarity:

  • most time series models

    • ARMA, ARCH, GARCH → ACF & PACF

    • Linear regression models on time series (OLS)

    • Bootstrap (if not stationarity → Block bootstrap)

    • VaR, ES


ARMA, ACF & PACF

Wold Theorem → building block of the ARMA process

Main idea: Any stationary time series YtY_{t} can be written as a linear function of current and past white noise ϵt\epsilon_{t} .

Lag operator Ljϵt=ϵt−jL^{j}\epsilon_{t}=\epsilon_{t-j}

Key formula:

ARMA Process

  • Instead of infinite terms, approximate with finite polynomials:


  • Leading to:

  • General ARMA(p,q) form:

  • AR(p): YtY_{t} regressed on past values → PACF cuts off after lag p

  • MA(q): YtY_{t} ​regressed on past shocks → ACF cuts off after lag q


ACF & PACF

  • ACF: Measures correlation between Yt​Y_t​ and Yt−τY_{t-\tau}

  • PACF: Measures correlation of YtY_t & Yt−τY_{t-\tau} after removing intermediate lags

    • In an AR(τ)AR(τ) model: PACF = coefficient on lag ττ

→ identification of p & q with ACF & PACF

Residuals analysis

  • After fitting a model, residuals should be white noise (stationarity)

  • Use Ljung-Box or Box-Pierce test:

→ Q(H) should be close to 0, IF residuals = white noise


ARMA Procedure

  1. Check stationarity (Dickey-Fuller test)

  2. Identify p & q from ACF (for MA) / PACF (for AR)

  3. Estimate ARMA(p,q) model

  4. Check residuals for white noise (Ljung-Box test)

    • RH0 (p-value < α\alpha ) : reject significant autocorrelation exists → model = inadequate

    • NRH0 (p-value >α\alpha ): no significant autocorrelation & residuals are white noise → model = well specified


Key rules

AR(p) → ACF tails off gradually

2 lags in PACF

MA(q) → ACF cuts off sharply after q

3 lags in ACF

ARCH & GARCH

AR models, variance (volatility) is assumed constant over time (homoscedasticity)

Reality: Financial series often show volatility clustering (periods of high/low volatility)

Solution: Model volatility to depend on past → ARCH model (Autoregressive Conditional Heteroscedasticity)

BUT when estimated on financial data, ARCH(p) usually needs large number of p of significant lags

Solution: GARCH (Generalized ARCH) model


2 - Step Procedure (ARMA(p1, q1) - GARCH(p2, q2))

Step 1 (mean)

  1. Use ACF/PACF plots to select p1 & q1

  2. Estimate ARMA(p1,q1) model

  3. Check residuals for white noise (Ljung-Box test + ACF/PACF of residuals)

    • RH0 (p-value < α\alpha ) : Residuals ≠ white noise, this means model missed some autocorrelation → model ≠ adequate

    • NRH0 (p-value >α\alpha ): Residuals = white noise, this means model captured the mean structure well → model = fits well

Step 2 (variance)

  1. Test for ARCH effects

  • ACF/PACF of squared residuals

  • Formal ARCH test: Regress et2e_{t}^2 on its lags, compute TR2TR^2 , compare to chi-squared

    • RH0 (p-value < α\alpha ) : Significant evidence of ARCH effects, this means the residuals show some time-varying volatility → GARCH model = needed

    • NRH0 (p-value >α\alpha ): No evidence of heteroskedasticity (volatility clustering), this means the model has constant variance → no need for GARCH

  1. Use ACF/PACF of squared residuals to select p2 & q2

  2. Estimate GARCH(p2,q2) model

  3. Check residuals after GARCH for remaining ARCH effects (compute ARCH test + ACF/PACF of ARMA(p1,q1) - GARCH(p2,q2) squared residuals)


ARMA-GARCH model

→ GARCH(p2, q2) can be rewritten as an ARMA model on squared Y’s errors (not directly on squared observations):

This form shows how past squared errors and volatility affect current squared error