
- •Contents
- •Introduction
- •Context
- •Graph 1 – Situation X da – company a’s demand dc – competitors’ demand
- •Financial Data Mining Literature
- •Picture 1 – Word ratio and stock performance visual comparison (Nagar and Hahsler 2012, pp. 15)
- •Competitors’ Effect Literature
- •Critical Analysis and Gap in Research
- •Proportion of competitors’ influence on the stock return
- •Proportion of the news’ effect on competitor(s)’ return
- •Justification
- •Industry return as control variable
- •Controlling for the market
- •Graph 3 – Agilent Technologies squared returns
- •Ratio Analysis
- •Competitors’ Influence - Regression
- •Industry return as control variable
- •Influence on Competitors – Regression
- •Regression analysis results
- •Industry return as control variable
- •Conclusion
- •Proportion of competitors’ influence on the stock return
- •Industry return as control variable
- •Proportion of the news’ effect on competitor(s)’ return
- •Possibility for practical use
- •Limitations of the research
- •Recommendations for future study
- •References
- •Appendix
Influence on Competitors – Regression
But our main question is whether significant news affects the competitors. In the previous regressions we had all the data, but now only significant news are analysed. Approximately 10% of the largest returns (both positive and negative) were chosen as “important” news (the number is equal to the number of outliers in the first tests). I regressed competitors’ returns on company A’s significant returns and controlled for the industry using S&P technology index.
– denotes significant returns at time t for company A
– is the constant term
– is the return of the first (second) competitor at time t. If the company A has only 1 competitor, the second regression is not calculated.
– is the industry return (return of the S&P index for technological companies only) at time t.
Proportion of the news’ effect on competitor(s)’ return
H0: company A’s significant return does not negatively affect its competitors’ returns on the same day.
Ha: company A’s significant return negatively affects its competitors’ returns on the same day.
|
Competitor 1 |
Competitor 2 |
|
Competitor 1 |
Competitor 2 |
||||
|
p-value |
|
p-value |
|
p-value |
|
p-value |
||
A |
0.2451 |
0.0073 |
0.2866 |
0.0000 |
KLAC |
0.3936 |
0.0000 |
|
|
AAPL |
-0.7736 |
0.0000 |
-0.9780 |
0.0000 |
LLTC |
0.6494 |
0.0000 |
0.6237 |
0.0000 |
ADI |
0.8126 |
0.0000 |
|
|
LXK |
0.1858 |
0.0134 |
|
|
ALTR |
0.3693 |
0.0000 |
|
|
MSFT |
-0.0485 |
0.1191 |
-0.0708 |
0.3635 |
AMAT |
0.4154 |
0.0002 |
|
|
MXIM |
0.2980 |
0.0000 |
0.3140 |
0.0001 |
AMD |
0.0253 |
0.2442 |
0.0659 |
0.0284 |
NCR |
-0.0117 |
0.8934 |
0.0427 |
0.1433 |
BRCM |
0.1678 |
0.0009 |
0.1680 |
0.0076 |
NTAP |
0.1025 |
0.3227 |
0.2772 |
0.0002 |
CIEN |
0.0669 |
0.2331 |
|
|
QCOM |
0.2036 |
0.0365 |
|
|
CSCO |
0.2601 |
0.0071 |
0.1812 |
0.0250 |
QLGC |
0.1453 |
0.0000 |
0.1446 |
0.0000 |
EMC |
0.2499 |
0.0041 |
0.0566 |
0.5656 |
SANM |
0.0458 |
0.1162 |
|
|
GOOG |
0.4007 |
0.1035 |
0.1565 |
0.0146 |
TMO |
0.2218 |
0.0006 |
|
|
HPQ |
0.0345 |
0.4132 |
|
|
TXN |
0.5289 |
0.0000 |
|
|
IBM |
0.1999 |
0.1549 |
0.1266 |
0.2976 |
XLNX |
0.9892 |
0.0000 |
|
|
INTC |
0.3791 |
0.0840 |
0.0490 |
0.3629 |
XRX |
0.1222 |
0.2233 |
|
|
|
|
|
|
|
YHOO |
0.3076 |
0.0226 |
0.0800 |
0.1349 |
Table
9 – Results
|| Estimated
values of
for
competitors 1 and 2
P-value
less than 0.1 denotes 90% significance level
P-value
less than 0.05 denotes 95% significance level
P-value
less than 0.01 denotes 99% significance level
This time the only significant results are shown by Apple. It has a dramatic opposite effect on its competitors. The results are also highly significant. NCR and Microsoft also show some opposite effect, but the results are insignificant and are very close to 0.
Discussion of the Results
Ratio analysis results
|
Competitor 1 |
|
Competitor 2 |
||||||
|
OR |
RN |
Neg |
OR |
RN |
Neg |
|||
AAPL |
45.33% |
5.26% |
54.39%* |
50.70% |
1.75% |
29.82% |
|||
EMC |
49.30% |
5.56% |
55.56% |
43.94% |
1.85% |
20.37% |
|||
LXK |
45.33% |
3.08% |
55.38%* |
|
|
|
|||
MSFT |
46.72% |
5.77% |
59.62%** |
48.71% |
5.77% |
51.92% |
|||
QLGC |
43.74% |
3.85% |
36.54% |
45.13% |
1.92% |
48.08% |
Table
10 – Abridged
table 6. Comparing Neg and RN ratios to the Opposite Ratio
*
- significant at 90% confidence level
** - significant at 95%
confidence level
If we recall the results from the ratio analysis, very few companies had Neg ratio higher than OR ratio. It means that significant news on average have less situation X component than news in general. Only Apple (AAPL) and its first competitor – Google (GOOG), Lexmark (LXK) and its competitor – Hewlett Packard (HPQ) and Microsoft (MSFT) and its competitor – Apple (APPLE) showed significant opposite relationships.
The research question asks if situation X occurs more often than situation Y, which means if the Neg ratio is significantly higher than 0.5. The results are:
|
Competitor 1 |
|
Competitor 2 |
||||||
|
OR |
RN |
Neg |
OR |
RN |
Neg |
|||
AAPL |
45.33% |
5.26% |
54.39% |
50.70% |
1.75% |
29.82% |
|||
EMC |
49.30% |
5.56% |
55.56% |
43.94% |
1.85% |
20.37% |
|||
LXK |
45.33% |
3.08% |
55.38% |
|
|
|
|||
MSFT |
46.72% |
5.77% |
59.62%* |
48.71% |
5.77% |
51.92% |
|||
QLGC |
43.74% |
3.85% |
36.54% |
45.13% |
1.92% |
48.08% |
Table
11 – Modified
table 10. Comparing Neg and RN ratios to the 0.5
*
- significant at 90% confidence level
Only Microsoft affects its competitor, Apple in opposite direction significantly more often than in 50% of the cases. Therefore computer trading programs can earn arbitrage profits.
To do so a program analyses Microsoft news articles. If they are important and positive, then the computer short sells Apple stock. If they are important and negative, then the computer buys Apple shares. The research suggests that approximately 59.62% of transactions will be.