Below, you’ll find my personal CFA level 3 applications of economic analysis notes…

You can find a list of the other categories here: **CFA Level 3 Notes, Formulas, and Weights.**

**Beta Research** is
related to returns from systematic risk. **Alpha
Research** is related to excess risk adjusted returns from a particular
strategy.

Number of distinct correlations in a matrix is the number of assets squared minus the number divided by two.

**Gross Domestic
Product** (GDP) measures production in a country and includes labor and
property that is both domestically and foreign owned. Components include:
Consumption (often makes up 60-70% of GDP), Government Spending, Capital
Spending by Businesses, Net Exports.

**Output Gap** is the difference between estimated GDP from trends and actual GDP. As the gap closes, inflation tends to rise. If real GDP goes above estimated GDP, the economy is under inflationary pressure.

**Gross National Product** (GNP) measures the market value of services and goods produced by all citizens of a country both abroad and domestically.

**Data Measurement
Errors and Biases**

**Transcription Errors**from collecting and recording data**Survivorship Bias****Appraisal (Smoothed) Data**tends to show a lower correlation with other assets and lower volatility than the real underlying.**Data-Mining Bias**is “drilling” data to find patterns that can’t be expected to have predictive value.**Time-Period Bias**

Technological, political (regime), legal and regulatory
changes can lead to **Nonstationary**
drifts. Long-term averages and trends might not persist.

**Exogenous Variables**
are determined outside of a system and **Endogenous
Variables** are determined within.

Correlation Doesn’t Necessarily = Causation

**Psychological Traps**

**Anchoring Trap****Status Quo Bias****Confirming Evidence Trap****Overconfidence Trap****Prudence Trap**is tempering forecasts so they don’t seem extreme.**Recallability Trap**

**Descriptive
Statistics** summarizes and describes data. **Inferential Statistics** makes estimates or forecasts about a larger
data set.

**Shrinkage Estimators**
is a weighted average of a covariance and another covariance estimator… reminds
me of Blume’s Beta.

**Volatility Clustering** is when big price swings are followed by more big swings. Past Volatility can persist and this formula tries to show the relationship…

**Gordon’s Growth Model** = D1 / P0 + g

**Grinold-Kroner Model**

**Repurchase Yield**
is the second component to the formula above.

**Build-Up Approach**
adds various risk premiums together to determine expected return.

**Bond-Yield-Plus-Risk-Premium
Method** adds a risk premium to long-term government YTM.

**Standard Error** =
Standard Deviation / Sqrt(Sample Size)

**International Capital
Asset Pricing Model** (ICAPM): E(R) = Rf + B[E(Rm) – Rf)

- ICAPM assumes perfect markets (no liquidity issues or transaction costs)

**Black-Litterman**
approach reverse engineers expected returns and incorporates an investor’s
views.

**Singer-Terhaar
Approach**

**Perfect Market Integration** implies that there are no barriers to capital movements cross borders. **Market Segmentation** implies there are barriers to capital movement.

If markets are completely segmented, the correlation component in the formula above equals one and can be omitted. This makes sense because a segmented market should have a higher risk premium.

**Singer-Terhaar Steps**

- Estimate both the perfectly integrated and perfectly segmented risk premiums
- Add illiquidity premiums (if any)
- Determine the degree of the assets integration
- Take a weighted average

Covariance of Two Assets = Product of the Betas Times Market Variance

**Taylor Rule**

**Permanent Income
Hypothesis** asserts that long-run income expectations determine spending
behavior

**Diffusion Indexes**
measure how many indictors point up and how many point down.

**Molodovsky Effect**
occurs when the P/E ratios of cyclical companies are above their historical
means during downturns when a recovery is expected.

**Forecasting Exchange
Rate Approaches**

**Purchasing Power
Parity** (PPP) asserts that exchange rate movements should offset inflation
rate differences between two countries.

**Relative Economic
Strength** looks at investment flows instead of trade flows.

**Capital Flows**
look at long-term flows such as equity investment and foreign direct investment
(FDI).

**Savings-Investment
Imbalances**

**Cobb-Douglas Production Function**

Total Factor Productivity aka Solow Residual

**H-Model Formula**

**Justified P/E** is
the valuation estimated dividend divided by the expected earnings.

Bottom-up estimates are often more optimistic (vs. top-down) heading into a recession and more pessimistic heading into a recovery. Top-down can be slow in detecting cyclical turns.

**Fed Model** asserts
that the S&P earnings yield (usually forward earnings yield) should be
equal to the 10-year treasury yield.

**Forward Earnings Yield with Gordon Growth Model**

**Yardeni Model**

Yardeni Model improves upon the Fed Model by incorporating some risk with a Moody’s A-rated corporate bond… and also using dividend over earnings.

**Cyclically Adjusted
P/E Ratio** (aka CAPE and Shiller’s P/E) takes the moving average of 10 years
of real (CPI adjusted) earnings.

**Tobin’s q** is the
market value of equities and liabilities (assets combined) dividend by the
replacement cost of its assets (include liabilities in numerator).

Tobin’s q > 1 : indicates further capital investment should be profitable

Tobin’s q < 1 : indicates further capital investment should be unprofitable

**Equity q** is the equity market cap dividend by net worth measured at replacement cost (subtract liabilities in denominator from assets at replacement cost). This differs from price-to-book value ratio since price-to-book uses historic values.

Please comment below if you have any suggestions or questions. Also, the next category in my CFA level 3 study list is **Asset Allocation**.

Invest mindfully,

Brian Kehm