Yahoo Inc - Statistics Project Example

I. Introduction This paper presents the trend analysis and forecast for Yahoo Inc. The company was founded in 1994 by two Stanford PhD candi s (Yahoo Inc, 2014). The company offers emails, entertainment, news and sports (Yahoo Inc, 2014). The current Chief Executive Officer of Yahoo Inc. is Marissa Mayer who is assisted by fourteen other members of the executive team (Yahoo Inc, 2014). The aim of this paper is to examine the performance of Yahoo Inc. and forecast sales, costs and profits for the next two years. The objectives of this paper are: 1. To identify the trends in sales, costs and profits of Yahoo Inc. from 2003 to 2012. 2. To forecast the performance of Yahoo Inc. in terms of sales, costs and profits for the next two years. From the analysis in this paper, it is found that Yahoo Inc. sales have been declining and the forecast sales for the next two years are $ 1,852,400 in the first year and $ 3,650,091 in the second year. The mean sales was $ 5,381,647 with a standard deviation of $ 1,722,645. The trend of Yahoo Inc. costs was quadratic in nature for the period under analysis. The mean cost for the current period was $ 3,822,612 with a standard deviation of $ 2,044,539. The predicted costs for the next two years are $ 1,622,362 for the first year and $ 2,111,307 for the second year. Finally, the results showed that profits of Yahoo Inc. followed a cubic trend with forecasted profits for the next two year being $ 1,113,059 and $ 773,138 respectively. The mean profit for the current period was $ 1,162,794 with a standard deviation of $ 1,082,864. II. Data and Methodology The data collected for this exercise is for Yahoo Inc. from 2003 to 2012 annual data giving a total of 10 year observations (see Table 1 for the raw data collected). Secondary data were collected from the annual statements of Yahoo Inc. freely available on their website (Yahoo Inc, 2003-2012). Secondary data refers to data collected for other purposes other than the present analysis. Since our aim is to assess the performance of Yahoo Inc. on various performance indices which are readily available in the form of secondary sources, we consider the use of these sources as they are also very reliable. It is important to observe trends in the time series data in order to assess which model to apply when undertaking a forecast of data. In this regard, graphical presentations are usually recommended. Thus in this study, graphs are used to show the trends in the data gathered as has been recommended by other scholars for observing time series data (Anderson, et al., 2010). One of the objectives of this paper was to assess the measures of central tendency and measures of dispersion (spread). In this study, we use mean scores and median values to assess central tendency and standard deviation, minimum and maximum values to assess spread in the time series data. Therefore, this has been done for sales, costs and profits of Yahoo Inc. for the period under review. These results are presented in table formats in Part II of this paper. There are many methods that can be used to forecast time series data such as the one in this study (Table 1). In our assessment of the task at hand and the limitation of resources, we conducted the forecast using regression analysis technique which we found to be appropriate for this study. There are a number of statistical softwares available for conducting forecasts of time series data (Evans, 2003). These include but are not limited to Excel, SPSS, Stata, Eviews, R, and Minitab. Again, due to resource limitations and our assessment of the task at hand, we use the Excel software to conduct the forecasting exercise in the present study. The results of the forecast are provided in part III of this paper and appendices 1-3 provide the excel outputs of these analyses. Table 1: Yahoo Inc. Sales, Costs and Profit Data 2003 – 2012. Year Sales Costs Profit 2003 1,625,097.0 959,339.0 237,879.0 2004 3,574,517.0 1,543,598.0 839,553.0 2005 5,257,668.0 2,053,742.0 1,896,230.0 2006 6,425,679.0 2,808,990.0 751,391.0 2007 6,969,274.0 3,435,103.0 660,000.0 2008 7,208,502.0 7,195,539.0 418,921.0 2009 6,460,315.0 6,073,623.0 597,992.0 2010 6,324,651.0 5,552,127.0 1,231,663.0 2011 4,984,199.0 4,183,858.0 1,048,827.0 2012 4,986,566.0 4,420,198.0 3,945,479.0 Figures 1, 2 and 3 present the trends observed in the data. As shown in Figure 1, sales took a polynomial trend (with an order of 2) rising from 2003 to 2008 and falling again to 2012. The trend of costs can be said to be linear as there is a general upward rise in the costs over the period of analysis (as shown in Figure 2). Figure 3 shows that the trend of profit for the period under analysis is also polynomial (of order 3) rising, falling and then rising. III. Analysis of Central Tendency and Spread III.1 Sales of Yahoo Inc. Table 2 presents the descriptive results on sales of Yahoo Inc. for the period 2003 – 2012. The mean sales value was $ 5,381,647 with a median sales of $ 5,791,160. The sales value deviated from the mean by $ 1,722,645. The lowest sales value for the company was $ 1,625,097 while the highest sales value was 7,208,502. Yahoo Inc. has seen a general decline in sales as was observed in Figure 1. Table 2: Summary Descriptive Statistics for Sales of Yahoo Inc. 2003 – 2012. Sales Mean 5,381,647 Median 5,791,160 Standard Deviation 1,722,645 Minimum 1,625,097 Maximum 7,208,502 Observations 10 III.2 Costs Table 3 shows the summary descriptive results for the costs (total operating expenses) of Yahoo Inc. for the period 2003 – 2012. The results show that Yahoo Inc. had a mean cost of $ 3,822,612 with a median cost of $ 3,809,481. The standard deviation of costs for the company was $ 2,044,539. The costs ranged from the lowest of $ 959,339 to the highest cost of $ 7,195,539. A linear trend of costs observed for the company in Figure 2 could mean that costs are likely to keep rising in the future. Table 3: Summary Descriptive Statistics for Costs of Yahoo Inc. 2003 – 2010. Costs Mean 3,822,612 Median 3,809,481 Standard Deviation 2,044,539 Minimum 959,339 Maximum 7,195,539 Observations 10 III.3 Profits Table 4 shows the descriptive results for the profits (net income) of Yahoo Inc. from 2003 – 2010. The mean profit was $ 1,162,794 and the median profit was $ 795,472. The standard deviation of profits was $ 1,082,864. The profits for the company ranged from a low of $ 237,879 to a high of $ 3,945,479. As was shown in Figure 3, there is a general rise in profitability of the company since 2008 and the rise in 2012 was a very sharp rise than in other years. This may suggest that the company will still be profitable in the future. Table 4: Summary Descriptive Statistics for Profits of Yahoo Inc. 2003 – 2010. Profit Mean 1,162,794 Median 795,472 Standard Deviation 1,082,864 Minimum 237,879 Maximum 3,945,479 Count 10 IV. Forecasting of Yahoo Inc. Sales, Costs and Profits IV.1 Yahoo Inc. Sales Forecast Given that the trend of sales was of a quadratic nature, a regression analysis was run of the following form: Sales = b0 + b1t + b2t2. According to Levine, Stephan, Krehbiel, & Berenson (2013), some forecasts involve quadratic regression and not usually linear regressions. The results of the regression yield to formulation of the following equation: Sales = -766,321,945,086 + 763,188,327 Year – 190015 Year2. The predicted sales values for the next two years are therefore $ 1,852,400 and $ 3,650,091 respectively. Appendix 1 shows the regression outputs for sales. IV.2 Yahoo Inc. Costs Forecast The trend of costs was of linear nature. Thus, a linear regression is estimated to forecast the costs of Yahoo Inc. for the next two years. The equation was of the following nature: Sales = b0 + b1t. The results show that the predicted costs for the next two years are $ 1,622,362 and $ 2,111,307 respectively. Appendix 2 shows the regression outputs for costs. IV.3 Yahoo Inc. Profits Forecast It was found that the trend of profits was of cubic nature from Figure 3. Thus, a cubic function was estimated using regression analysis. The function was of the following nature: Profit = b0 + b1t + b2t2 + b3t3. The predicted profits for the next two years were $ 1,113,059 and $ 773,138 respectively. Appendix 3 shows the regression outputs for profits. V. Comparison of Individual Results A comparison between the findings on sales, costs and profits for Yahoo Inc. is made. It is found that as the sales and profits values are declining for the period under review, the costs are steadily rising. Sales take a quadratic trend while profits take a cubic trend. On the other hand, the trend of costs is linear. A comparison of the forecast methods shows that all the forecasts are based on regression analysis. For the sales forecast, a quadratic equation is used and then the forecasts for the next two years predicted from the resulting equation. This results in sales that are lower than the mean predicted for the next two years. For the costs forecast, a linear regression model is estimated. The resulting equation is then used to predict costs for the next two years. This also results in costs lower than the mean. Further, a cubic equation is formulated and tested and the resulting equation is used to predict profits for the next two years. As the results show, the resulting forecasts are also below the mean. A comparison of forecasted results reveal a similar trend – all the results found for the next two years fall below the current mean values. Sales values are $ 1,852,400 and $ 3,650,091 for the next two years, costs are $ 1,622,362 and $ 2,111,307 for the next two years while profits are $ 1,113,059 and $ 773,138. From the comparisons above, Yahoo Inc. is doing badly in all the three trends at least as at the moment given the dataset provided and the forecasted values for the next two years. Sales are declining and so is the profits while the costs are steadily rising. The rising costs are yet to continue and the fluctuations in sales and profits will still be observed. VI. Conclusion and Recommendations The Yahoo Inc. is not doing well in terms of sales. Revenues are declining and the predictions show that sales of $ 1,852,400 and $ 3,650,091 will be made in the next two years. This is way below than the highest sales figures ever recorded for the period under review ($ 7,208,502) and also below the mean sales for the period studied ($ 5,381,647). Profits are also forecasted to be below the mean in the next two years with predictions being are $ 1,852,400 and $ 3,650,091 while the current mean of profit is $ 1,162,794. The results also show that the costs are rising and the forecast for the next two years is $ 1,622,362 and $ 2,111,307 which is still below the mean cost of $ 3,822,612. This study faced a number of limitations. First, a 10 year observations is used in the study to make the forecasts. While this was enough to conduct the forecast, the resulting results cannot be relevant enough for an accurate prediction of future performance. It would be important to conduct a further analysis with a large dataset for sales, costs and profits. Secondly, the study uses annual data to conduct the forecasts. This data does not provide the kind of seasonality and randomness required to observe better trends and therefore correctly predict future performance. In future, it would be important to use quarterly data to capture these fluctuations in performance. Third, the study uses regression analyses methods to forecast performance of sales, costs and profits. This method does not do well in accurately forecasting such data given the assumptions of the model. Time series decomposition methods are recommended in future in order to calculate the trends and then employ regression analysis on the decomposed data to perform forecasts. We did not test the data for the assumptions of regression analysis as is always the recommendation of most statisticians (Ramsey & Schafer, 2013). Therefore, the data we ran may have given spurious results as far the forecasts are concerned especially when time series data is used (Cornwell, 2009; Lim, 2011). It would therefore important for future studies to check for the assumptions of normality, heteroskedasticity, and non-stationarity of time series data. There are numerous softwares available for conducting forecasting of time series data but we chose Excel. This software has been found to have a number of limitations (Goldwater, 2007) and therefore future research should employ other sophisticated software such as Eviews, R, or Stata for conducting such an analysis especially if the dataset is large. An analysis made in this study focuses on only one firm – Yahoo Inc. There is therefore no basis upon which to judge whether the results on performance as far as sales, costs and profits are better or worse. If in the future data could be gathered for other peer firms or the industry results available, it would be worthy to conclude whether Yahoo Inc. is doing well or worse than the peers in the industry. The study makes a number of recommendations for Yahoo Inc. to consider. Currently, the level of sales are declining and this could be due to the high competition in the technology market at the moment. In order to improve sales, it is important that the company thinks about more innovative ideas and products to offer to the market as this would improve the current levels of sales. While the company is not currently making loses, the rising costs are reducing the profit levels of the firm. It is important that the rising costs be investigated to keep them at minimum as they eat into the revenues generated by the firm. A thorough audit of these costs would provide a basis upon which an objective decision can be made on how to reduce the costs. References Anderson, D. R., Sweeney, D. J., Williams, T. A., Camm, J. D., Cochran, J. J., Fry, M. J., & Ohlmann, J. W. (2010). Quantitative Methods for Business. Ohio: South-Western Cengage Learning. Cornwell, D. (2009). An Exploratory Time Series Analysis of Apprehensions and Linewatch Hours on the South West Border. Washington: Office of Immigration Statistics. Evans, M. K. (2003). Practical Business Forecasting. MA: Blackwell Publishers Ltd. Goldwater, E. (2007, February). Using Excel for Statistical Data Analysis - Caveats . Retrieved from University of Massachusetts School of Public Health : http://people.umass.edu/evagold/excel.html Levine, D. M., Stephan, D. F., Krehbiel, T. C., & Berenson, M. L. (2013). Statistics for Managers Using Microsoft Excel. London: Pearson. Lim, K. G. (2011). Financial Valuation and Econometrics. London: World Scientific Publishing. Ramsey, F., & Schafer, D. (2013). The Statistical Sleuth: A Course in Methods of Data Analysis. Boston: Brooks/Cole Cengage Learning. Yahoo Inc. (2003-2012). Annual Report. California: Yahoo Inc. Yahoo Inc. (2014, May 2). What Do We Do. Retrieved from Yahoo Website: https://info.yahoo.com/about-us Yahoo Inc. (2014, May 2). Who Are We. Retrieved from Yahoo Website: https://info.yahoo.com/about-us Yahoo Inc. (2014, May 2). Yahoo Executive Team. Retrieved from Yahoo Website: https://info.yahoo.com/management-team Appendices Appendix 1: Sales Excel Regression Output SUMMARY OUTPUT Regression Statistics Multiple R 0.975598662 R Square 0.951792749 Adjusted R Square 0.938019248 Standard Error 428868.5735 Observations 10 ANOVA   df SS MS F Significance F Regression 2 2.54E+13 1.27E+13 69.10319 .000 Residual 7 1.29E+12 1.84E+11 Total 9 2.67E+13         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Intercept -766321945086 7.52E+10 -10.2 .000 -9.442E+11 -5.9E+11 Year 763188327 74936413 10.2 .000 585991869 9.4E+08 Yearsq -190015 18664.11 -10.2 .000 -234148.74 -145882 RESIDUAL OUTPUT Observation Predicted Sales Residuals 1 1,852,400 -227303 2 3,650,091 -75574.4 3 5,067,752 189915.9 4 6,105,383 320296.4 5 6,762,983 206291.1 6 7,040,553 167949.2 7 6,938,093 -477778 8 6,455,602 -130951 9 5,593,081 -608882 10 4,350,530 636035.9 Appendix 2: Costs Excel Regression Output SUMMARY OUTPUT Regression Statistics Multiple R 0.724052024 R Square 0.524251333 Adjusted R Square 0.46478275 Standard Error 1495754.764 Observations 10 ANOVA   df SS MS F Significance F Regression 1 1.97E+13 1.97E+13 8.815602 0.017896 Residual 8 1.79E+13 2.24E+12 Total 9 3.76E+13         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Intercept -977733088.8 3.31E+08 -2.95754 0.018215 -1.7E+09 -2.2E+08 Year 488944.3091 164677.2 2.969108 0.017896 109198.1 868690.6 RESIDUAL OUTPUT Observation Predicted Costs Residuals 1 1,622,362 -663023 2 2,111,307 -567709 3 2,600,251 -546509 4 3,089,195 -280205 5 3,578,140 -143037 6 4,067,084 3128455 7 4,556,028 1517595 8 5,044,972 507154.5 9 5,533,917 -1350059 10 6,022,861 -1602663 Appendix 3: Profits Excel Regression Output SUMMARY OUTPUT Regression Statistics Multiple R 0.699986 R Square 0.48998 Adjusted R Square 0.201403 Standard Error 876878.8 Observations 10 ANOVA   df SS MS F Significance F Regression 3 5.17E+12 1.72E+12 3.362482 0.096197 Residual 7 5.38E+12 7.69E+11 Total 10 1.06E+13         Coefficients Standard Error t Stat P-value Lower 95% Upper 95% Intercept 8.83E+10 5.13E+10 1.722608 0.128627 -3.3E+10 2.1E+11 Year 0 0 65535 #NUM! 0 0 Yearsq -65829.4 38161.36 -1.72503 #NUM! -156067 24407.89 Yearcub 21.87657 12.67291 1.726247 0.127948 -8.0901 51.84323 RESIDUAL OUTPUT Observation Predicted Profit Residuals 1 1113059 -875180 2 773138 66415.02 3 564602.5 1331627 4 487583.4 263807.6 5 542211.9 117788.1 6 728619.2 -309698 7 1046937 -448945 8 1497295 -265632 9 2079827 -1031000 10 2794662 1150817 Read More