perils of using ols to estimate multimedia communications effects abstract companies invest millions dollars in various forms marketing impact customers awareness attitudes purchases and ultimately profitability an important question for marketers shareholders alike is what do investments have on market performance assess these mix models by regression analysis however we show that the estimation via ie ordinary least squares yields severely biased estimates specifically our examples overstates media advertising sales 34 147 other words estimated effect can be twice as much it really synergy carryover are under 28 48 respectively mitigate such severe biases present alternative approach called wiener kalman filter which provides reasonable closer true parameters than corresponding addition analyze corolla brand s campaign furnish results based marketplace data corroborate simulation findings finally discuss both implications managers opportunities
