![]() CATEGORIES: BiologyChemistryConstructionCultureEcologyEconomyElectronicsFinanceGeographyHistoryInformaticsLawMathematicsMechanicsMedicineOtherPedagogyPhilosophyPhysicsPolicyPsychologySociologySportTourism |
Competitors’ Influence - RegressionHowever, 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.
- - - - - 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, Market return as control variable H0: there is no negative relationship between the return of company A and its competitors. Ha: there is negative relationship between the return of company A and at least one of its competitors.
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
- RAt – denotes the return at time t for company A - Β0 – is the constant term - RAt-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. - Rcomp1(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. - RspTECHt – 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. Ha: there is negative relationship between the return of company A and at least one of its competitors.
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: 847
|