Competitors’ Influence - Regression

However, this testing method may not take into account the proportions of the effects given by competitors and the market. To analyse these proportions I used a regression involving the return of the stock, its previous day return, return(s) of its competitor(s) and return on the market.

- – denotes the return at time t for company A

- – is the constant term

- – is the previous day return for company A, implies AR(1) form for the regression. Its coefficient, B₁ shows the effect of the previous day’s return on the current day’s return.

- – is the return of the first (second) competitor at time t. If company has only 1 competitor, the second value is not included into the regression.

- – is the market return (return of the S&P 500 index) at time t.

The idea behind this regression is to try to find the effect of each of the competitor on the firm A’s return. To control for market-wide news another regressor, is added. represents daily returns of the Standard and Poor Index. The Beta-Coefficients for the competitors’ returns would show their effect on the company A, and if the coefficients are negative we would conclude that there is some opposite relationship between competitors’ returns. In this regression all the data is included (2 years of observations).

Market return as control variable

H₀: there is no negative relationship between the return of company A and its competitors.

H_a: there is negative relationship between the return of company A and at least one of its competitors.

	Competitor 1	Competitor 2		Competitor 1	Competitor 2
	P-value		P-value		P-value		P-value
A	0.1360	0.0005	0.3802	0.0000	KLAC	0.4778	0.0000
AAPL	0.1137	0.0604	0.0395	0.2576	LLTC	0.6186	0.0000	0.2678	0.0000
ADI	0.4332	0.0000			LXK	0.1905	0.0000
ALTR	0.8326	0.0000			MSFT	-0.0361	0.2818	-0.0297	0.5095
AMAT	0.3904	0.0000			MXIM	0.5645	0.0000	0.2522	0.0000
AMD	0.4933	0.0001	0.1049	0.4614	NCR	0.0097	0.7612	0.0949	0.1954
BRCM	0.3247	0.0000	0.3299	0.0000	NTAP	0.0378	0.2202	0.5959	0.0000
CIEN	0.4670	0.0000			QCOM	0.1425	0.0010
CSCO	0.0851	0.0034	0.1204	0.0000	QLGC	0.1445	0.0465	0.1662	0.0028
EMC	0.0304	0.2519	0.4357	0.0000	SANM	0.4540	0.0000
GOOG	0.0082	0.6543	0.0850	0.0521	TMO	0.2431	0.0000
HPQ	0.2582	0.0118			TXN	0.2408	0.0000
IBM	0.0567	0.0923	0.0484	0.0130	XLNX	0.3646	0.0000
INTC	0.0635	0.0001	0.1369	0.0027	XRX	0.1294	0.0000
					YHOO	0.0362	0.0900	0.1151	0.0562

Table 7 – Results || Estimated values of and 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

Again, only Microsoft shows negative relationship with its competitors. However, the relationship is insignificant. Moreover, most companies show highly significant positive relationship with its competitors. As I mentioned above, the correlation may lie in the industry-wide news that please both the company A and its competitors. I therefore ran another regression, but controlling for industry instead, rather than the market. This time, daily returns of iShares Standard and Poor's Gsti Technology Index Fund are used as the control variable and are denoted by :

- R^A_t – denotes the return at time t for company A

- Β₀ – is the constant term

- R^A_t-1 – is the previous day return for company A, implies AR(1) form for the regression. Its coefficient, B₁ shows the effect of the previous day’s return on the current day’s return.

- R^comp1(2)_t – is the return of the first (second) competitor at time t. If company has only 1 competitor, the second value is not included into the regression.

- R^spTECH_t – is the industry return (return of the S&P index for technological companies only) at time t.

Industry return as control variable

H₀: there is no negative relationship between the return of company A and its competitors.

H_a: there is negative relationship between the return of company A and at least one of its competitors.

	Competitor 1	Competitor 2		Competitor 1	Competitor 2
	p-value		p-value		p-value		p-value
A	0.2620	0.0000	0.5433	0.0000	KLAC	0.6159	0.0000
AAPL	-0.1225	0.0000	-0.1480	0.0000	LLTC	0.6531	0.0000	0.2636	0.0000
ADI	0.6169	0.0000			LXK	0.2218	0.0000
ALTR	0.8655	0.0000			MSFT	-0.5681	0.0000	0.0561	0.2091
AMAT	0.5132	0.0000			MXIM	0.5983	0.0000	0.2475	0.0000
AMD	0.7195	0.0000	0.3132	0.0241	NCR	0.0436	0.2060	0.3416	0.0000
BRCM	0.3368	0.0000	0.4525	0.0000	NTAP	0.0450	0.1442	0.6212	0.0000
CIEN	0.5312	0.0000			QCOM	0.2329	0.0000
CSCO	0.1325	0.0000	0.0869	0.0028	QLGC	0.1969	0.0004	0.2870	0.0000
EMC	0.0196	0.4525	0.5082	0.0000	SANM	0.4169	0.0000
GOOG	0.0157	0.4281	0.1982	0.0000	TMO	0.4254	0.0000
HPQ	0.3557	0.0001			TXN	0.4520	0.0000
IBM	0.0631	0.0023	0.1650	0.0000	XLNX	0.4520	0.0000
INTC	0.1026	0.0000	0.2480	0.0000	XRX	0.1928	0.0000
					YHOO	0.0445	0.0534	0.2693	0.0000

Table 8 – Results || Estimated values of and 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

Similar situation happens even if we control for technology industry instead of the market as a whole. The only difference is that Apple showed highly significant negative relationship with its competitors and Microsoft’s opposite relationship with the first competitor became much stronger and highly significant. It however showed no more negative effect from the second competitor.

Date: 2015-12-17; view: 793