Economic Forecasting - Research Paper Example

? Economic Forecasting Economic Forecasting Economic Forecastingis a process by which businesses predict the future economicsituation. It involves the use of statistical models that utilizes variables called indicators. Some of the economic indicators used include retail sales and unemployment rates, GDP decline / growth, inflation, and interest rates(Elliott, Granger, and Timmermann, 2006).Economic forecasting is an important decision-making tool for governments and businesses as they project on the financial strategy and policy. Annualbudgeting and forecasting cannot keep pace with today’s vibrant business situation because the information produced is irrelevant andoutdated. Therefore, financial managers should be capabletoidentify and respond promptly to the impact of competitive forces and changes affecting their businesses. Most organizations fail to forecast on the fiscal impact of these changes fast enough; thus, affecting their businesses. Periodical forecasts take two to five weeks to finalize, and budgets are normally not finalized until the actual year of budgeting. The time taken to produce each repetition of the budget or forecast is too long taking days and sometimes weeks, and in the current environment, the impact of any change to the financials need to be understood within a short period(Elliott, Granger, and Timmermann, 2006).It is startling that the need for faster delivery of advanced forecasts and budgets has not received more consideration more so in light of the time and drive toward a faster close Most budgeting and forecasting processes and systemsrequireacceptablechanges to accommodate thereorganizations, mergers, acquisition, and divestitures that have become the characteristic of modern businesses. This creates historical comparisons and trend analyses that old questionable value. In addition, most systems are not flexible enough to accommodate the demand for multiple views of forecast and budget information. Nonetheless, some of the elements that need to be considered in the economic forecasting are discussed herein below. 1) Time Series Plot of sales This pattern exists when a series is influenced by seasonal factors includingdays of the week, months and the quarters of the year. There are usuallycumulative trend and seasonal peak between the index andindex 15. Several methods of analysis are possible, for example: 1. Exponentially smoothing and Forecasting Convenient that allows frequent information 2. Season /Trend decomposition is Simple and easy to communicate 3. Time series estimate and modelingis the most advanced and is not easily communicated 4. Regression with seasonal models, flexible and Intermediate Simple season or trend decomposition does not pick up possible monthly auto-correlation as the other methods do. Regression with seasonal and trend modelshasadded opportunity of groupingtogether months of comparable sales level to give a model with few factors and with potentially better forecasts. 2)Descriptive Statistics: Sales Variable N N* Mean SEMean StDev MinimumQ1 MedianQ3 Maximum Sales 16 0 5370 257 1030 3896 44605288 6010 7707 Autocorrelation Function: Sales Lag ACF T LBQ 1 0.058206 0.23 0.07 2 0.181312 0.72 0.74 3 -0.001481 -0.01 0.74 4 0.552103 2.13 8.06 Quantity connection function is obtained when the quantity in high-level item is alteredwith the quantity of sub-item. When thisis done, the change in the proportion is realized automatically, and the Sales orderSales ACF linkage and delivery group can be updated manually that is defined in the above graph that in turn defines the following conditions: 1. Sales ACF linkage existed without delivery group then the extent correlation function is active. 2. Delivery group existed without sales ACF linkage then the extent correlation function is inactive. 3. Sales ACF linkage existed with delivery group then the extent correlation function is active. If the delivery Sales ACF linkage is updated manually then delivery group cannot be updated then, 1. Sales ACF linkage existed without delivery group then the extent of the correlation function is inactive 2. Sales ACF linkage existed with delivery group then the extent of the correlation function is active. 3. Delivery group existed without sales ACF linkage then the extent of the correlation function is inactive. When the major difference inthe extent of thecorrelation function isinactive in delivery if no delivery group maintain even Sales ACF linkage has happened. 3. Trend Analysis for Sales Data Sales Length 16 NMissing 0 Fitted Trend Equation Yt = 4187 + 139*t Accuracy Measures MAPE 11 MAD 632 MSD 581882 Appropriate identification of newly developing trends is very vital to businesses. Sales arrays of customer segments indicate market inclinations. Downward and Upward trends in sales indicate new market trends. Time-series projecting model can be used to identify new market trends fixed in changes of sales returns. Understanding of sales trends is important for customer retention as well as for marketing. 4. Time Series Decomposition Plot for Sales A Time Series Analysis Method forms the basis of forecasting on the hypothesis that the near future will depend on the past and that any past arrangements will continue. Time series analysis is concerned with the identification of the model that best fits the past data.Since the best method of discovery, this model depends on the shape taken by the previous data. Time series of past data Suitable methodBest fitting model Forecast;therefore,the first stage is to produce a scatterplot against time leaving room for guesses in the near future. 5. Component Analysis for Sales For a sales environment to be successful, there are four key structural elements that must exist including: 1. Goals-For sales to remain actual and these goals must be well-defined goals. 2. Skills of the sales team; the team must have the necessary skills required for selling and managing the sales situation. 3. Process-there must be a procedure that maps sales from lead origin sequentially to the final deal. 4. Tracking- there must be tracking systems that can identify the current position of indications in the sales process and how long they take at each stage of the process. 6. Seasonal Analysis for Sales For the regression method,the initial stage iscreating a variable representing the time trend, and generating indicator variables dummyto indicate each month of the year. As projected by the above graphs, there is a constructive time trend indicating that all months have constructive regression constants (Elliott, Granger, and Timmermann, 2006). This implies that the basis month January is the lowest season. Moreover, there are indications that 89.7% of the disparity in sales is defining the TIME TREND and the monthly indicators. Additionally, the remaining standard deviation of S=3121 gives an estimate of the prediction errors. Reference Elliott, G., Granger, C. W. J., &Timmermann, A. (2006).Handbook of economic forecasting. Amsterdam: Elsevier North-Holland. Read More