 # NYB Harvard Case Solution & Analysis

## NYB Case Study Analysis

Method-2: Moving Average

Under moving average, forecasting is done through taking average of previous values taking a particular number in the simple average of values formula in excel.. In this case, the moving average is based on 4 historical values and is calculated as follows:

Total bike Pick-ups = (350+176+210+267)/4

Total bike Pick-ups (March 1, 12am-1am) = 251

Error % = 46%

Method-3: Naïve Approach

Under Naïve Approach the forecasted value is based on the previous value, which means the forecasted value of total bike pick-ups will be the value of total bike pick-ups on Feb 28 during 12am-1am that is 267.

Total bike Pick-ups (March 1, 12am-1am) = 267

Error % = 49%

All of the above methods could be evaluated and compared on the basis of the % error under each of the method. From the following table 1, it could be seen that the Linear Regression has the least percentage error among all of the three methods. Although, the total bike pickups under each method has a slight difference, but the Linear Regression provides more accurate results, with least % Error.

Table-1

 Summary Pick ups Error % Linear Regression 261 43% Moving Average 251 46% Naïve Approach 267 49%

Suggested Forecasting Method

From the above analysis, the suggested forecasting method is Linear Regression with the least percentage error and high chances of accuracy. Linear regression is considered to be the most accurate method as it considers complete historical data with moving average considering only a limited time data and naïve approach only considering the previous value. Therefore, Linear Regression could be considered to be an optimal forecasting method and the forecasted value for March 1 under this method, are provided in Appendix-2.....................................

This is just a sample partical work. Please place the order on the website to get your own originally done case solution.

Other Similar Case Solutions like