cross asset classes

overtime

i.e. explanatory variables of return

record mistakes

poor performance data is deleted

overestimate return

hedge fund

inliquid infrequent priced asset -> smoothier returns

estimate risk

e.g. real estate

avoid regime change -> nonstationary data

fulfill sample size requirement

more precise estimate

use more frequent data -> asynchronous data

over-exploit the temporary relationship

mitigation: economic sense? examine modeling process? test with out-of-sample data

short-term focused

long-term focused

Structural Model

Reduced-Form Model

incorporate multiple variables

can reuse

quantified based on consistent set of relationship

complex and time-consuming to construct

difficult to forecast changeable relationship

unrealistic output difficult to interpret

does not work will to forecast turning points

leading indicators

coincident/lagging indicators

simply, intuitive, and easy to interpret

readily available from third party

can be tailored to meet specific forecasting needs

inconsistent forecasting result

economic indicator may give false signals

revised frequently -> fit past data better than now

less complex than econometrics

flexible with mixed objectives

subjective

time-consuming

complexity limited due to manual process

The growth rate of inflation has slowed down, but it is still the growth of inflation.

harms

Quantitative Easing (QE)

Cash

Bonds

Stock

Real estate

For details, please see Notebook Supplement Golden Process Online Course

not stimulative or restrictive

It seems that change is the main factor

confidence high -> dif. small

confidence low -> dif. large value lose -> to compensate risk

sell before maturity

Changable interest rate

interest rate fall

interest rate rise

interest rate fall

interest tate rise

calculate from most liquid instruments

close to gov. zero-coupon yield central bank policy rate

major factors

other indicators

risk higher

strategy to take advantage

highest of new issurance bond

issue at/close to par or market rate

new

large in size

issued by frequent and well-known issuer

simple in structure

high credit quality

wealth concentration

dominance of cyclical industries, e.g. commodity and less pricing power

capital flow and trade restrictions

inadequate fiscal and monetary policy

poor work force infrastructure week tech advancement

large foreign currency borrowing

less developed small capital markets

exposure to volatile capital flows

e.g. structural form, policy, government is also important

growth rate = nominal growth in GDP = real g inflation

features

E(R) = D/P (% delta(E) - % delta(S)) % delta(P/E)

expected return = dividend yield (exp. % change in total earnings) ( exp. % change in share outstanding)(share repurchase) exp. % change in the P/E ratio

3 components

NOI growth rate close to GDP growth rate in the long-run

NOI is a nominal measure

assumption: infinite time period

assumption: finite time period

long-term asset

not pay tenant

fluctuation in real estate value, leases, vacancy

impact of smoothed data

illquidity premium

consider local real estate nature -> not global

e.g. REIT -> benefit from diversification

reasons of current account structural imbalance

Hot Money

strong eco. country's currency -> more supply in global market -> weaken the current higher risk premium

mitigations

No. of observations >> 10 * No. of portfolio assets

advantage: consistent unfounded

disadvantage: large sample required -> N(N-1)/2

advantage

disadvantage

weighted average estimate of the above 2 methods

the above 2 methods' mitigation