← Back to input·Analysis Result
예시 포트폴리오 — 미국 분산 (AAPL · MSFT · SPY · TLT)
●Fama–French 3F · Portfolio personality
The Luxury Director명품 디렉터
Holding the classic identity of luxury intact, you read the flow of trends with a flexible, tasteful eye.
Regression coefficients · annualized
α (Alpha)Not sig.
-1.08%
t = -0.34p = 0.735
Almost fully explained by factors (alpha ≈ 0).
Market beta (βM)✓ Sig.
0.83
t = 14.36p = 0.000
Moves about 17% less than the market.
Size beta (βS)✓ Sig.
-0.20
t = -2.30p = 0.025
Weakly co-moves with large-caps when they lead.
Value beta (βH)✓ Sig.
-0.26
t = -4.05p = 0.000
Weakly co-moves with growth stocks when they lead.
Regression equation
Regression
Rp − Rf = α + βM·MKT + βS·SMB + βH·HML + ε
Sample: last 70 months of monthly data · OLS regression · values annualized
Annual expected return
+13.7%
The expected return reflects exposure to risk. Expected losses rise proportionally with expected return.
* Statistical estimate · not a guarantee of realized return.
Activation Guide
Guide: 5 ways to use the 3-Factor analysis result
FF3 · Quick guide
Check which risks your portfolio is exposed to. Adjust the portfolio strategically to take the risk exposures you want.
Visualization
Fama–French 3-factor exposure
My portfolioMarket (MKT only)
The closer to a corner, the stronger the exposure to that factor. Negative ≠ bad — negative SMB/HML simply means "tilted toward large/growth" — a style diagnosis, not a quality verdict.
Interpretation
3 takeaways
FactorComposite tilt — 2 factors
2 factors break threshold at mild-to-clear strength (size beta -0.20 (large-cap tilt) · value beta -0.26 (growth tilt)). No single axis dominates, but the book carries several colors at once. If that's by design, fine. If it's accidental, it's a good moment to decide which tilt you actually want to keep.
Alphaα -1.08% — effectively zero
Excess return beyond factor exposure is essentially nil (α -1.08%), and it's statistically indistinguishable from zero (p=0.735). Nearly all your performance is being explained by factor exposures — market, size, value, and the rest. That's neither good nor bad; it's a clean factor-driven book.
StabilityR² 79.8% — strong fit
The factor model explains 79.8% of the variation in your returns — a strong fit, typical of reasonably diversified equity books. The familiar axes (market, size, value) are doing a solid job of describing your big picture.
Factor Target
Set factor targets → optimized weights
Move the sliders to define a goal like "I want a market beta of 0.7" — we propose a weight mix that satisfies the target.
Market beta target
βM
+0.83→+0.83
+0.0+1.6
Size beta target
βS
-0.20→-0.20
-1.0+0.6
Value beta target
βH
-0.26→-0.26
-1.1+0.5
Rebalance intensityBalanced (moderate)
Next step
Want to separate quality and investment-style effects too?
Fama–French 5-Factor adds the RMW (profitability) and CMA (investment) factors and estimates profitability beta and investment beta separately.
FAQ
Frequently asked questions
Questions newcomers to this model commonly ask. Click any question to expand the answer.
Can I trade exactly the way the analysis suggests?
No, that is not recommended. Every analysis here is a statistical estimate based on historical market data, and it does not guarantee future returns or losses. Trading costs, taxes, FX costs, market impact, and your own investment goals and horizon are also not reflected. Use the results as a diagnostic of "what kind of risk and return profile this portfolio has", and before any real investment decision review your own situation and consider consulting a qualified professional.
I re-ran the analysis on the same tickers but the result is slightly different — why?
That is expected. Each run fetches the latest prices from yfinance and recalculates. When new trading days arrive or the factor data is updated between runs, the sample changes and the means, variances and regression coefficients shift slightly. The risk-free rate used for the Sharpe ratio (the US 13-week T-bill yield) is also refreshed daily. If the change is large (e.g. alpha flips sign, or a beta moves more than 0.3), the sample may be too short for stable results — check the analysis reliability score at the top of the result page.
How are Korean and US stocks handled together?
Holding amounts are converted to KRW at the FX rate at analysis time to set the weights. Returns, however, are calculated from each holding's local-currency price, so the returns and volatility of US holdings do not include KRW/USD exchange-rate moves. If you invest in won, your actual account can swing more or less than shown, by the amount the exchange rate moves. Markowitz and HRP analyze holdings from both markets as one portfolio. Fama–French (3- / 5-factor) regresses Korean holdings on DIVA's own Korea factors and US holdings on factors from the Kenneth French Data Library, then combines the two by portfolio weight into a single result.
How is the "analysis reliability" score calculated?
A 0–100 measure of how much the numbers could shift because of limited data or chance (not whether the portfolio is good). For Markowitz and HRP it averages ① the risk estimate (data window, and window length relative to the number of holdings) and ② the return estimate (the 95% error range of the past average return), so the same portfolio over the same window scores the same in both. Fama–French averages the beta estimate (window and beta error range) and explanatory power (R²). 75+ reliable, 55+ worth referencing, 35+ read conservatively, below that interpret with care. A heuristic for reference, not an academic standard.
Alpha is positive but reported as not statistically significant. What does that mean?
Alpha came out positive, but the regression cannot tell whether it reflects a lasting excess return or chance. Significance is usually claimed when the p-value is below 0.05 (roughly |t-stat| > 2). The p-value is the probability of an estimate this far from zero even if the true alpha were zero; above 5%, a true alpha of zero cannot be ruled out. Short windows or volatile returns can leave alpha looking large while significance stays low. Run the same portfolio through the 5-factor model to see whether the alpha is explained by the profitability and investment factors.
Does a positive size beta (β_S) mean my portfolio is full of small-caps?
No. Size beta measures whether your portfolio's returns co-moved with small-cap strength during the window — not what is inside the basket. So a holding classified as small-cap can still show a size beta near zero or negative if it moved like a large-cap over the window, and the reverse can happen too. The same logic applies to value, profitability and investment beta — these are all "co-movement patterns of returns", not labels of holdings. The label of what you hold and the factor exposure of its returns are separate concepts.
If R² is low, should I distrust the analysis?
More precisely, "read the result more conservatively" rather than "distrust it entirely". R² shows how much of the portfolio's return variance is explained by the three factors. Above 60% is the typical range for a well-diversified portfolio. Below 40% means idiosyncratic variance dominates, and the confidence intervals on beta and alpha widen — the same beta could move by ±0.3 or more. Take signs and rough magnitudes of beta / alpha as the takeaways, not exact numbers.
Why is alpha (α) left out of the expected-return breakdown?
The annual expected return on screen is "current risk-free rate + each factor beta × that factor's average premium over the analysis window". Alpha is the return left over after the factors, and it is usually not statistically distinguishable from zero, with little evidence that it repeats in the next period, so it is not counted in the forward-looking expected return. Alpha is shown separately, with its significance, in the regression coefficients at the top. Factor premiums are historical averages over the window, so if that period was unusually strong or weak, the expected return leans the same way.
Where does the Fama–French factor data come from? Is it free?
US-stock factors come free from the official data library run by Kenneth French at Dartmouth's Tuck School of Business (US factors for the 3-factor model, North American factors for the 5-factor model). They are usually published a month or two late, so the latest month may be missing from the regression. Korean stocks use DIVA's own KRW factors, computed from every KRX stock including delisted ones (from July 2016). DIVA calculates and adds new months separately, so recent months not yet added are left out of the Korean regression. Mixed KR / US portfolios are regressed on the two factor sets separately and combined by weight. Source links are in References at the bottom of the Fama–French model explanation page.