Page 19 - FINAL CFA I SLIDES JUNE 2019 DAY 4

LOS 12.e: Distinguish between a statistical result and an
economically meaningful result, p283 Session Unit 3:
12. Hypothesis Testing
Statistically significant result Economically significance considerations!
Say you reject Ho and hence support Ha • Transaction costs, Taxes, Risk
that 20 year returns to a strategy are +ve.
Do you execute the strategy?
LOS 12.f: Explain and interpret the p-value as it relates to hypothesis testing

Say a 2TT at 5% SL P-value is the smallest SL for which the hypothesis would be rejected.

The p of obtaining a test statistic that would lead to a rejection of Ho,
assuming it is true!

Reject Ho Reject Ho Lets interpret p-Values –in the context of
Meaning what?
Per Z-table, p. value > Z (= 2.3) = hypothesis testing of coefficients:
(1 – 0.9893) = 1.07%.
We compare p-value to SL…
Fail to Reject Ho

z =
= 2.3 SL = 2.14%- SL = 2.14%+
1.96
(1.07%- each side) < p-value = 2.14% (1.07%+ each side)
Independent variable or coefficient (1.07% each side)
is statistically speaking, NOT significantly different from zero: < Independent variable or coefficient
Fail to Reject Ho is statistically-speaking, significantly different
from zero,
Reject Ho
2TT, so 2 × 1.07 = 2.14%:
What is this? p-value = 2.14%
(1.07% each side)
At a 3% (1.5% for 2TT), 4%, or 5% SL, we would <
Its called p value!
Reject Ho, but at a 2% or 1% SL we would not. SL = 5%
(2.5% on each side)