Class Notes (1,017,905)
US (397,845)
WKU (174)
ECON (39)
Strow (13)
Lecture 1

ECON 203 Lecture Notes - Lecture 1: Statistical ModelPremium

8 pages36 viewsSpring 2017

Department
Economics
Course Code
ECON 203
Professor
Strow
Lecture
1

This preview shows pages 1-2. to view the full 8 pages of the document.
Regression for Options 25% ITM
for_reg=subset(results, (UnderlyingPrice-Strike)/UnderlyingPrice>(0.0)
& (UnderlyingPrice-Strike)/UnderlyingPrice<(0.25)
& Last>0.01
& exp_payoff<20
& Last<20
& actual_payoff<25)
Benchmark Model- How well does the market predict payoff (Last vs. Actual)
Call:
lm(formula = actual_payoff ~ Last, data = for_reg)
Residuals:
Min 1Q Median 3Q Max
-21.950 -2.417 -0.426 2.455 14.989
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 2.99303 0.17989 16.64 <2e-16 ***
Last 0.99776 0.02684 37.17 <2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 4.292 on 1860 degrees of freedom
Multiple R-squared: 0.4262, Adjusted R-squared: 0.4259
F-statistic: 1381 on 1 and 1860 DF, p-value: < 2.2e-16
R2 = 42.62% - as we can see from the graph, the market does not predict the
payoff accurately based on the last price of the option.
find more resources at oneclass.com
find more resources at oneclass.com
You're Reading a Preview

Unlock to view full version

Subscribers Only

Only half of the first page are available for preview. Some parts have been intentionally blurred.

Subscribers Only
How well the expected payoff mimics the Last price (Exp_payoff vs Last)
Call:
lm(formula = Last ~ exp_payoff, data = for_reg)
Residuals:
Min 1Q Median 3Q Max
-7.6133 -0.6888 -0.0075 0.6858 10.5237
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 0.717374 0.074559 9.622 <2e-16 ***
exp_payoff 0.780406 0.009946 78.461 <2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 1.786 on 1860 degrees of freedom
Multiple R-squared: 0.768, Adjusted R-squared: 0.7678
F-statistic: 6156 on 1 and 1860 DF, p-value: < 2.2e-16
R2 = 76.8% The data points fit the trendline much better in this example. The
option price generated by the model fits the market price quite well.
find more resources at oneclass.com
find more resources at oneclass.com
You're Reading a Preview

Unlock to view full version

Subscribers Only

Only half of the first page are available for preview. Some parts have been intentionally blurred.

Subscribers Only
Does expected payoff add anything to the predictive power of Last?
Call:
lm(formula = actual_payoff ~ Last + exp_payoff, data = for_reg)
Residuals:
Min 1Q Median 3Q Max
-21.5852 -2.1865 -0.3599 2.2896 14.5518
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) 2.43017 0.17170 14.15 < 2e-16 ***
Last 0.25014 0.05211 4.80 1.72e-06 ***
exp_payoff 0.75973 0.04641 16.37 < 2e-16 ***
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
Residual standard error: 4.014 on 1859 degrees of freedom
Multiple R-squared: 0.4985, Adjusted R-squared: 0.4979
F-statistic: 923.8 on 2 and 1859 DF, p-value: < 2.2e-16
with a p-value of 2e-16 (or 2 with 15 zeros in front of it), the exp_payoff
variable is significant in predicting payoffs.
When we include expected payoff, the value of our R2 increases by 7.23% (from
42.62% to 49.85%) This roughly translates into expected payoff being able to
predict payoffs about 7% better than the market.
find more resources at oneclass.com
find more resources at oneclass.com
You're Reading a Preview

Unlock to view full version

Subscribers Only

Loved by over 2.2 million students

Over 90% improved by at least one letter grade.