Scribd
Upload
81 views6 pages

QM Notes

1) The document discusses various quantitative methods for evaluating investment projects and portfolio performance including net present value (NPV), internal rate of return (IRR), time-weighted return, money-weighted return, and risk measures like standard deviation. 2) It also covers topics like yields, returns, averages, distributions, probability, and statistical tests that are important for quantitative analysis. 3) Finally, it mentions techniques for technical analysis of market trends, patterns, and indicators.

Uploaded by

api-294072125
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
81 views6 pages

QM Notes

1) The document discusses various quantitative methods for evaluating investment projects and portfolio performance including net present value (NPV), internal rate of return (IRR), time-weighted return, money-weighted return, and risk measures like standard deviation. 2) It also covers topics like yields, returns, averages, distributions, probability, and statistical tests that are important for quantitative analysis. 3) Finally, it mentions techniques for technical analysis of market trends, patterns, and indicators.

Uploaded by

api-294072125
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

Quantitative Methods

Accept proj w IRR > than firms reqd IRR; When NPV contradicts IRR -> Choose proj. w higher NPV!
NPV method assumes cfs are reinvested @ Opportunity cost of capital; this approach represents mkt
based cost of capital and is reqd rate of return for shareholders of the firm.
IRR method assumes cfs reinvestment rate is IRR.
HPR = end-beg / beg (or end/beg-1)
Money Weighted = IRR of portfolio (use the CF calc method)
Time weighted -> measures compound growth rate (break it down into periods *geometric return)
Money weighted gives more weight to period w more money in account. If funds added to portfolio just
before period of poor performance, MW < TM. If funds added just before period of high returns,
MW>TW. TW is a better measure of managers ability to select investments. MW is better measure if
manager has complete control over inflows and outflows of money into account.

Band Discount Yield: BDY = Discount/face value X 360/days (simple interest; FACE V not purch price)
HPY: end beg / beg OR (end/beg) 1 maybe?
% discount = fv-price / FV*
EAY: [(1 + HPY)365/t ]- 1 (annualized valueaccount for compound interest. )
MMY: = HPY X (360/days)
BEY: 2 X effective semi-annual yield OR HPY X (365/days) .* Double check this for certain
EAR = (1 + Periodic Rate) m - 1
Geometric Mean Return:

Harmonic Mean =

Harmonic Mean < Geometric Mean < Arithmetic Mean

Population Variance:

Pop Stnd. Deviation

Sample Variance:

Sample Stnd. Deviation

Mean Abs deviation =

Chebyshevs inquality: 1 1/k2 = % of distribution that lies within


Coefficient of Variation = CV = sx / x

= of x / avg value of x

Measures risk per unit of return

Nominal Risk Free Rate = real rate + expected inflation


Required rate = nominal risk free rate + default premium + liquidity premium + maturity risk premium
Scales:
Nominal = Random; Ordinal = ordered; Interval = Relative ranking; Ratio = ranking w equal differences
Sharpe Ratio = rp rf / p (bigger is better)

Distributions:
Symmetry -> deviations of ive and +ive are just as likely (mean = median = mode)
Skewness refers to the extent that distribution is not symmetrical.

+ive skewed : mean> median > mode

-ive skew: mean < median < mode

Kurtosis measures peak


Leptokurtic = more peaked; platykurtic = less peaked; meso = same as normal
Sample skewness & Kurtosis = (same as variance except each deviation is 3 and 4 instead of 2)

Covariance = measure of how 2 assets move together (range ive to +ive infinity)
Correlation coefficient = measures strength of linear relationship (ranges from 1 to +1); = covariance /
product of two s (NOT Variances 2)

Port Return:

CoVariance

Portfolio Variance:

Probability:
Empirical probability = past data; priori = formal reasoning/inspection; subjective = personal judgement
Odds = p occurring / (1-p occurring)
Multiplication rule (A and B)

Addition rule (A or B)

Pg 205 onwards important stuff!!!

Bayes Formula (pg 220*)

Order Patters:

Order does not matter:

nPr = n! / (n-r)!

nCr = n! / (n-r)! r!

binomial distribution:
p(x) = [ n! / (n x)! x!] * p x * (1-p) n x

probability distribution = sum of the prob. of all outcomes add to 1


discrete random variable = outcomes can be counted
continuous random variable = cannot be counted (virtually infinite and p of any one = 0)
tracking error = difference between portfolio return and a benchmark
time-series data = over a period of time; cross sectional data = at a single point in time
Normal Distribution pg 253:
U +/- z

z = (observation pop mean) /

Standard error = x = /

=xu/

Confidence intervals: z/ /

OR

Normal Dist. w known variance


Normal Dist. w unknown variance
Non - Normal Dist. w known variance
Non - Normal Dist. w unknown variance

z/ S/
Z
T
n/a
n/a

Z
T
Z
T

Test statistic = (sample statistic hypothesized value ) / standard error of sample statistic
Type 1 error = rejecting null when its actually true : p (type 1) = signif level (eg. 5%)
Type 2 error = not rejecting the null when its actually false: Power of a test = 1- P (type 2 error)

Confidence Intervals
{ sample stat (crit. value) (std. error)} pop. parameter { sample stat (critical v) (/ )}
- critical value test statistic + critical value
T-test: difference of means; must be independent; must be normally distributed
Paired comparisons: test whether means of differences between observations between two
samples are different
T = - d2 / 2
Chi-Squared (X2): test concerning variances of 2 Normally distributed populations
X2 = (n-1)2 / 2
F-Distribution: testing equality of Variances of 2 normally distributed populations based on 2
independent random samples.
F = S12 / S22 (S12 is the larger one, numerator is larger)

Technical Analysis:
Downtrend: lower lows and lower highs

Up trend : higher lows and higher highs

Breakout and breakdown

Support and resistance

Change in polarity -> breached resistance becomes support and vice versa
Reversal Patterns
Head and Shoulders and Inverse Head and Shoulders
Double and Triple Tops and bottoms
Moving averages; short-term crosses long term.
Bollinger Bands;

Contrarians

Oscillators: Rate of change: diff. between latest closing and N periods


Relative Strength: rate of increases to decreases
Moving averages; Stochastic Oscillation
Put Call Ratio: Put / Call Volume increases are negative for value????
Elliot Wave Theory: up 5 waves, down 3 waves
Fibonacci Numbers: add the previous two numbers to produce the next. Ratios higher over lower?

You might also like