Quantitative and Qualitative Forecasting : Assignment
Added on - 21 Apr 2020
Quantitative and Qualitative Forecasting1QUANTITATIVE AND QUALITATIVE FORECASTINGNameCourse NumberDateFaculty Name
Quantitative and Qualitative Forecasting2Quantitative and Qualitative ForecastingExample 13.2: Comparing Simple and Weighted Moving Average, Exponential Smoothing andLinear Regression AnalysisTable1: Insurance company claim form process controlSampleNumber InspectedNumber offormsCompletedIncorrectlyFractionDefectiveSimpleMovingAverageWeightedMovingAverageExponentialSmoothing1300100.0333#N/A#N/A230080.02670.03000.03330.0333330090.03000.02830.02830.03304300130.04330.03670.02960.0329530070.02330.03330.03990.0334630070.02330.02330.02750.0329730060.02000.02170.02440.03248300110.03670.02830.02110.03189300120.04000.03830.03280.03201030080.02670.03330.03820.032412345678910-0.00500.01000.01500.02000.02500.03000.03500.04000.04500.05000.05500.06000.0650Fraction DefectiveLinear (Fraction Defective)Lower LimitUpper LimitSimple Moving AverageWeighted Moving AverageExponential SmoothingFigure1: Claim form control chart
Quantitative and Qualitative Forecasting3Figure 1 above shows various line graphs for claim form control chart, representingdiffering seasonality effects. The control chart displays the lower and upper limits, linearregression best fit line, exponential smoothing, simple & weighted moving averages and theoriginal scatter plot for the time against the defection fraction. Comparing the five plotted lines,the seasonality effects decreases significantly from the original line through simple movingaverage, weighted moving average, exponential smoothing and finally the linear plot which doesnot have any form of defects and seasonal effects (Brockwell and Davis, 2016). Based on thestated order, the power of time series analysis decreases significantly between these line charts,which the original plot depicting the best scenario and the linear regression the worst scenario tomeasure seasonality. However, as the plot is smoothened, an analyst is able to summarize andclearly understand the trend, hence acquiring powerful administrative information (Al-Omari andAl-Nasser, 2011). According to the linear regression line, there is a general reduction indefective proportions from the first to the tenth sample. Also, based on the exponentialsmoothing, the trend decreases significantly with minimal seasonal effects (Ahangar andChimka, 2015).Quality Management – ToyotaWasher thickness1.The probability of having washer above 2.4mm if the data is normally distributed.Count above 2.4mm = 0The count of washers above the thickness of 2.4mm in the sample means that there will be nowasher above the threshold (Lawrence, Klimberg and Lawrence, 2009).2.The proportion within 1.4 – 2.4mm range.