CFA LEVEL II 2025 FORMULA SHEET

F in Q uiz F ormula S heet CFA P rogram L evel II

2. Adjusted R2 5. Akaike’s information criterion (AIC)
QUANTITATIVE METHODS 𝑛−1 '')
"!
𝑅 =1−& ) (1 − 𝑅! ) AIC = n ln D -
E + 2(k + 1)
𝑛−𝑘−1

3. F-Statistic or F-Test 6. Schwarz’s Bayesian information

Learning Module 1: !"# %& '(")*+ *+,*+''-%. criterion (BIC or SBC)
&'( ( )
Basics of Multiple Regression and = = /
'')
&') (
!"# %& '(")*+' +**%*
) BIC = n ln D E + ln(n)(k + 1)
Underlying Assumptions .0/01 -

where,
df numerator = k = 1 Learning Module 3:
1. Multiple Regression
df denominator = n – k – 1 = n – 2 Model Misspecification
Yi = b0 + b1X1i + b2X2i + … + bkXki + εi,i =
1, 2, … n.
4. ANOVA
2. Prediction Equation 1. Breusch–Pagan (BP) test
Prediction equation = Y "! = b%" + b%# X#! + ANOVA SS MSS F Test statistic = n × R2residuals
b%$ X$! +. . . +b%% X%! + ε! , i SSR
' SSR- where,
Regression SSR k
= &(y)& SSE-
df = 1 k
&() (n − k − 1)
− y*)* R2residuals = R2 from a second regression
Learning Module 2:
of the squared residuals from the first
Evaluating Regression Model Fit and SSE
' regression on the independent variables
Interpreting Model Results Error SSE
= &(y& n = number of observations
df = n-2 n−k−1
&()
− y))*
2. Variance Inflation Factor VIF/
1. Coefficient of Determination: R2
SST 1
Sum of square regression - VIF/ =
= Total 1 − R$/
Sum of square total = A(y!
∑ni=1(Y= − Y>)2 df = n-1
!.#
= n
∑i=1(Yi − Y>)2 − y>)$

www.finquiz.com All rights reserved 1

 CFA LEVEL II 2025 FORMULA SHEET

Learning Module 4: yt = e b0 +b1t

7. Using Dickey-Fuller Test
Extensions of Multiple Regression
xt - x t-1 = b0 + (b1 -1) x t-1 + εt
3. Autoregressive Time-Series Models
First order autoregressive AR (1): 8. Smoothing Past Values with n-Period
xt = b0 + b1 x t-1 + εt
1. Studentized residual t !∗ Moving Average
pth-order autoregressive AR (p):
𝑡0 ∗ =
13∗
= N
6 3454#
5 xt = b0 + b1 x t-1 + b2 x t-2 + …..+ bp x t-p +εt xt + xt -1 + xt -2 + ..... + xt -( n -1)
24∗ 226(#47 33 )413
n
4. Mean reverting level 9. Correcting Seasonality in Time Series
2. Cook’s distance D!
𝑒0$ ℎ00 b0 Models:
𝐷0 = W Y xt =
𝑘 × 𝑀𝑆𝐸 (1 − ℎ00 )$ 1 - b1 For quarterly data
xt = b0 + b1x t-1 + b2x t-4 + εt
3. Linear regression
5. Chain Rule of Forecasting:
with 3 independent Variables
For monthly data
𝑌" = 𝑏# + 𝑏$ 𝑋$" + 𝑏! 𝑋!" + 𝑏% 𝑋%" + 𝑒"
One-period ahead forecast xt = b0 + b1x t-1 + b2x t-12 + εt
4. Logistic regression (logit) x̂t+1 = b̂0 + b̂1 xt
8 10. ARCH model =
= In (#48)
Two-period ahead forecast= eˆ 2 t = a 0 + a1eˆ 2 t -1 + µ t
= b0 + b1X1 + b2X2 +b3X3 + ε
# x̂t+2 = b̂0 + b̂1 xt+1
P = #91:;[4(=#># 9 =$>$ 9=?>? 9 @)]
where

6. Random Walks and Unit Roots: µt is an error term

Learning Module 5:
Time Series Analysis
Random Walk without drift: Predicting variance of errors in period
xt = x t-1 + εt where, b0 = 0 and b1 = 1. 2 2
t+1 = σˆ t+1 = α̂ 0 + α1εˆt
1. Linear Trend Models
yt = b0 + b1t+ εt Correcting Random Walk:
Learning Module 6:
yt = xt - x t-1
Machine Learning
Predicted/fitted value of yt in period (T +
1) Random walk with a drift:
xt = b0 + x t-1 + εt where, b0 ≠ 0 and b1 = 1
= yˆ t +1 = bˆ0 + bˆ1 (T + 1) 1. LASSO:
By taking first difference Penalty term (when l > 0) = 𝜆 ∑B5.#[𝑏>5 [
2. Log-Linear Trend Models yt = xt - x t-1 = b0 + εt

www.finquiz.com All rights reserved 2

 CFA LEVEL II 2025 FORMULA SHEET

3 B J8
TPR = JK9P8 1 + i∫
A(𝑌0 − 𝑌0 )$ + 𝜆 A[𝑏%B [ F∫ /Z = S∫ i j
`
Z (1 + iZ )
0.# 5.#
4. Root Mean Square Error: RMSE:
(8D1I0QR1I3 4SQRTFG3 )5 Using day count convention:
When l = 0, RMSE = ∑30.# 3 ' ! Actual $*
)1+ id # ,=
LASSO penalized regression = OLS ( " 360 &%+
regression ECONOMICS ' ! Actual $*' 1 *
S f /d )1+ i f # ,) ,
( " 360 &%+)( Ff /d ,+
Learning Module 7:
Big Data Projects Learning Module 1
Currency Exchange Rates æ é Actual ù ö
ç1+ i f ê ú÷
1. Normalization Ff / d = S f /d ç ë 360 û ÷
ç é Actual ù ÷
𝑋0 − 𝑋E03 1. Bid-offer Spread ç 1 + id ê ÷
𝑋0(3CDEFG0H1I) =
𝑋EF: − 𝑋E03 è ë 360 úû ø
Offer price – Bid price

where Xi = value of observation 2. Forward Rate: 6. Uncovered Interest Rate Parity :

Fwd rate = Spot Exchange rate i f - %DS e f / d = id
Performance Metrics: •
Forward points
2. Accuracy +
10,000 %DS e f / d = i f - id
J89JK •
=
LM 9 NM 9 LO 9 NO
3. Forward Premium or Discount: Forward premium or discount:
F1 score = (2*P*R)/(P + R) UVWXYWZ [V!-\]
spot exchange rate − ( ) • For one year horizon =
#","""
= Ff /d − S f /d =
spot exchange rate
where
−1 "i −i %
T = true, F = false, P = positive,
N = negative
S f /d $ f d ' ≅ S f /d (i f − id )
4. To convert spot rate into forward quote: # 1+ id &
Spot exchange rate × (1 + % premium) • Using day count convention:
3. Receiver Operating Characteristic (ROC):
Spot exchange rate × (1 - % discount) ( " Actual % +
* $ -
False positive rate: FPR # 360 '& -
5. Covered interest rate parity: Ff /d − S f /d = S f /d * (i f − id )
P8 * 1+ i " Actual % -
FPR = JK9P8
1 * d$
# 360 '& ,
-
(1 + iZ ) = S∫ g1 + i∫ h i j )
`
Z F∫ /Z
True positive rate TPR:

www.finquiz.com All rights reserved 3

 CFA LEVEL II 2025 FORMULA SHEET

7. Forward discount or premium as % of

Learning Module 1
spot rate: 5. Cobb-Douglas Production Function
Economic Growth
Ff / d - S f / d F (K, L) = Kα L1 - α
@ (i f - id )
S f /d 1. Economic growth: 6. Under Cobb-Douglas production
If uncovered interest rate parity holds Annual % ∆ in real GDP or in real per function:
• Ff /d − S f /d capita GDP
= = %ΔS ef /d ≅ (i f − id ) Marginal product of capital = MPK = α
S f /d 2. Relation between Stock Prices and AK α-1 L 1-α = α Y/K
Economic growth:
8. Purchasing Power parity (PPP) α Y/K = r èα = r (K) / Y = Capital
• Pf = S f/d × Pd )
P = GDP DbcME D)E
M income / Output or GDP
• S f/d = Pf / Pd
7. Output per worker or Average labor
where,
9. Relative version of PPP productivity (Y/L or y):
P = Aggregate value (price) of equities
= %∆S f/d = πf – πd GDP/Labor input = TFP × capital-to-labor
E = Aggregate corporate earnings
ratio × share of capital in GDP
10. Ex ante version of PPP Or
3. Expressing in terms of logarithmic rates:
= %∆Sef/d = πef – πed y = Y/L = Akα
• (1/T) % ∆P = (1/T) % ∆GDP + (1/T)
11. Real Exchange Rate %∆ (E / GDP) + (1/T) % ∆(P / E) 8. Contribution of Capital Deepening
• % ∆ in stock MV = % ∆ in GDP + % ∆ = Labor productivity growth rate – TFP
æ S f / d Pd ö æ ö
ç ÷ = S f / d ç Pd ÷ in share of earnings (profit) in GDP +
ç P ÷ çP ÷ % ∆ in the P/E multiple
qf/d = è f ø è f ø 9. Contribution of Improvement in
technology
4. A two-factor aggregate production Labor productivity growth rate – Capital
or
function: Deepening
æ CPI d ö Y = AF (K, L)
qf /d = S f /d ç ÷
ç CPI ÷ 10. Growth Accounting based on Solow
è f ø Y = Level of aggregate output in the Approach
12. Fisher effect: economy ∆Y /Y = ∆A / A + α ∆K/K + (1 – α) ∆L/ L
id = rd + πεd L = Quantity of labor or number of
if = rf + πεf workers or hours in the economy 11. Labor productivity growth accounting
if – id = (rf – rd) + (πεf- πεd) K = Stock of capital used to produce equation
(rf – rd) = (if – id) - (πεf- πεd) goods and services
A = Total Factor Productivity (TFP)

www.finquiz.com All rights reserved 4