【正文】 hs ? Deflators ? Regression analysis Supply side analysis ? Cost structures ? Design differences ? Factor costs ? Scale, experience, plexity and utilization ? Supply curves Demand side analysis ? Customer understanding segmentation and “Discovery” conjoint analysis multidimensional scaling ? Pricevolume curves and elasticity ? Demand forecasting technology/substitution curves Wrapup DEFLATORS CORRECT EFFECTS OF INFLATION Converts Variables from “Nominal” to “Real” Time series data in dollars with high or widely fluctuating inflation rates distort picture of growth Deflating data removes some of the distortion Using a deflator index list, currency data are multiplied by the ratio of the base year deflator index to the data year deflator index, ., ? 1979 sales (1993 $) = 1979 (1979 $) x Deflator 1993 Deflator 1979 SELECT APPROPRIATE DEFLATOR DEPENDING ON THE QUESTION YOU’RE TRYING TO ANSWER . deflator is best for expressing dollars in terms of average real value to the rest of the economy ? Current (variable) weights ? Measured quarterly . is best only for expressing value in relation to consumer spending on a fixed market basket of goods (1973 base) ? Measured monthly Industry or productspecific indices are best for converting dollars into measures of physical output ? Available from Commerce Dept. for broad industry categories ? Can be constructed from client or industry data for narrow categories BE CAREFUL WHEN MIXING EXCHANGE RATES AND INFLATION ACROSS COUNTRIES First convert each country’s historical data to constant local currency ? ., Japan—1993 yen ? W. Germany—1993 DM ? .—1993 dollars Then convert to single currency (dollars, for example) at fixed exchange rate EXAMPLE: AN INTEGRATED CIRCUIT MANUFACTURER Reported Sales . Deflator Average . Average . Year ($M) (1987 = ) Price ($) Transistor Price (162。 80% 70% 60% 50% 40% 30% 20% 10% 0% Annual Number of Purchases by Consumer X: Annual number of purchases by buyer Y: Percent ACV 185。 IRI Marketing Factbook。 sign of slope (here positive) indicates whether effect is positive or negative R2 = Multiple R R Square (%) Adjusted R Square (%) Standard Error Observations 29 Regression Statistics Regression 1 Residual 27 Total 28 Analysis of Variance df Sum of Squares Mean Square F Significant F Intercept () () () () X1 Coefficients Standard Error t Statistic Pvalue Lower 95% Upper 95% Sources: Scantrack。 BCG Analysis Microsoft Excel Regression Output HOW STRONG IS RELATIONSHIP? ‘ tstatistic’ measures how well X explains Y ? Simply calculated as slope divided by its standard error ? Closer slope is to zero, and/or higher standard error (variability), the weaker the relationship A shortcut: tstatistic greater in magnitude than 2 means relationship is very strong (., roughly 95% confidence level). Between and 2, relationship is relatively strong (., roughly 8595% confidence level). Under , relationship is weak. Multiple R R Square (%) Adjusted R Square (%) Standard Error Observations 29 Regression 1 Residual 27 Total 28 Regression Statistics df Sum of Squares Mean Square F Significance F Intercept () () () () x1 Coefficients Standard Error t Statistic Pvalue Lower 95% Upper 95% Analysis of Variance HOW WELL DOES MY MODEL WORK OVERALL? R2 measures proportion of variation in Y that is explained by the variables in the model here just X ? Indicates overall how well model explains Y ? Based on how dispersed the data points are around the regression line R2 measured on scale of 0 to 100% ? 100% indicates perfect fit of regression line to the data points ? Low R2 indicates current model does not fit the data well suggests there are other explanatory factors, besides X, that would help explain Y Multiple R R Square (%) Adjusted R Square (%) Standard Error Observations 29 Regression 1 Residual 27 Total 28 Regression Statistics df Sum of Squares Mean Square F Significance F Intercept () () () () x1 Coefficients Standard Error t Statistic Pvalue Lower 95% Upper 95% Analysis of Variance USE MULTIPLE REGRESSION TO SORT OUT EFFECTS OF SEVERAL INFLUENCES Use ? When several factors have an impact simultaneously ? To help distinguish cause from correlation Don’t use as “fishing expedition” MULTIPLE REGRESSION CAN ENHANCE PREDICTIVE ABILITY 0%20%40%60%80%10 1 0 0 1 0 0 00%20%40%60%80%0% 1 0 % 20% 30% 40%0%20%40%60%80%1 2 3 4 1 0 %0%10%20%30%40%50%60%70%80%0% 20% 40% 60% 80%% ACV with Features and/or Displays Brand Size Percent of Households Buying Annual Number of Purchases per Year % ACV with Features and/or Displays % ACV with Features and/or Displays Brand Size ($M) Percent of Households Buying Annual Number of Purchases/Year R178。=.51 R178。=.87 Predicted % ACV with Features and/or Displays Actual % ACV with Features and/or Displays Brand Size, Reach, and Purchase Freqency Sources: Scantrack。 BCG Analysis OTHER REGRESSION EXAMPLES 0%10%20%30%40%50%60%70%80%0 50 100 150 200 250 300 350 5%0%5%10%15%20%25%30%35%40%0 20 40 60 80 100 120 1400204060801001201 9 2 0 1 9 3 0 1 9 4 0 1 9 5 0 1 9 6 0 1 9 7 0 1 9 8 0 1 9 9 0 20231010010001920 1930 1940 1950 1960 1970 1980 1990 2023Very Low R178。 Certified Price Book。 BCG Analysis ** Source: Agricultural Statistics R178。=.002 R178。Additional functionality assumed 178。 8 100 100 — 16 36 53 18