Forecast for EBBD Company - Essay Example

Forecast/EESC Problem situation Forecasting refers to the prediction of future performance based on the present or past data. EBBD Company lacks credible financial planning due to the use ineffective methods. However, the company needs to carry out future forecasts that incorporate time series and inflation rates (Brandimarte & Zotteri, 2007). Assumptions The company neglects inflation rates during its financial planning. The company lacks efficient qualitative and quantitative data from its key stakeholders to make short and medium range financial decisions. The company employs extensive financial resources in forecasting of its financial obligations. EBBD will have a peak distribution demand during the unforeseen future which necessitates early planning by the management. The company lacks seasonal forecasting methods in its financial planning. The underlying factors during forecasting are neglected leading to deviations in the forecasted variables. The company lacks effective judgmental methods in evaluating its unique or new market conditions. The company lacks forecasting accuracy standards to analyze its forecast values against its actual values (Brandimarte & Zotteri, 2007). Evidence Simple Time series methods: Cumulative mean The technique can be used by EBBD where the demand is known based on the stationary process. Naive forecast The company can employ the technique to forecast demand based on prior observation using the noise term. The technique is non stationary meaning it has no specific pattern. The technique can be employed by middle level managers to make short term financial forecasts (Wang & Wang, 2010). Simple moving average The EBBD management can forecast a product’s demand if it has a long and flat pattern which leads to a permanent change or shift in the process. This may be caused by price changes or shift in competitors (Wang & Wang, 2010). Weighted moving average The management of EBBD can employ the technique where historical observations gradually lose their predictive value as opposed to the moving average which has an abrupt predictive value. Current data will provide forecast accuracy due to the upgrade of information. Simple exponential smoothing A forecast on future demand is arrived at by forecasting on the next period based on the weighted average between the last forecast and the last observation. It continuously updates the forecast series by the incorporation of current forecast errors. The technique inflates the forecast error while data remains stationary (Wang & Wang, 2010). Adaptive response rate exponential smoothing (ARRES) The technique employs the basic forecasting idea by avoiding parameterization. The technique responds to forecast errors where the large values are instituted as large and the small values as small eliminating the unbiased errors(Wang & Wang, 2010). Time series regression The technique incorporates the trended demand data so as it fits the historical data into the linear model. The model decomposes demand data observations into a trend component, initial level and noise components which were errors in the regression estimates (Wang & Wang, 2010). Holt’s procedure It forecasts demand data using a simple linear trend. The technique separates the temporary level from the trend data and thereby developing smooth estimates of every component. The appropriate values are developed based the trial and error method to minimize the MSE. Smoothing with seasonal indices In logistics and distribution, fluctuating customer demand is common. The smoothed estimates should are updated exponentially while observing every seasonal correction term. Seasonal indices are instituted for adjustment of forecasts (Wang & Wang, 2010). Winter’s model for seasonal and trended demand In logistics and distribution, demand patterns are exhibited by items which incorporate seasonality and trend. The model combines Holt’s procedure and seasonal index model to forecast on seasonality, level and trend. Every component is forecasted with individual smoothing coefficients and exponential smoothing (Wang & Wang, 2010). Qualitative and quantitative methods The top level managers at EBBD should institute qualitative forecasting techniques for long term financial projections. The qualitative methods are subjective in nature meaning they rely on consumers and experts opinions. They are mostly efficient when there is lack of prior information especially regarding to EBBD new markets. Some of the techniques used in qualitative methods include Delphi method, the historical life cycle analogy and market research techniques. Quantitative methods are used to forecast future statistics based on the past information. The quantitative methods will be employed by EBBD middle level managers to make short term decisions. This is because they are mostly efficient when past data is available to make future similar predictions. Techniques employed herein are simple exponential smoothing, N period moving averages and the multiplicative seasonal indexes (Wang & Wang, 2010). Causal or relational forecast method The concept of this method is that provides a relationship between the explanatory variables and the existing variables so as they can be inferred on the new market. The techniques used herein are: Life cycle model The distribution company can employ the model to analyze the phases of the product life. The technique evaluates the consumer behavior so as to analyze the demand pattern (Mukherjee & Kachwala, 2009). Simulation model The EBBD managers should incorporate the model so as to portray the distribution channel of the products in meeting the customers demand. They evaluate the time-lag effects and their dependent demand schedules. Time series methods The methods are the most accurate since they incorporate the SKU level and local dis-aggregated forecasts in the distribution system (Mukherjee & Kachwala, 2009). Forecasting accuracy Forecasting accuracy is the difference between forecast value and the actual value for the specific period. The company should incorporate measures like the Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) to forecast errors. Forecasting accuracy are recommended because it excludes forecasting errors that will in return cost the company in relation to time and costs (Mukherjee & Kachwala, 2009). Quarterly inflation forecasts models: UC-SV (stochastic volatility) The model provides that inflation fluctuates under un determined trend. The model employs standard deviation as means to respond to the temporary departures from the actual figure. In this model the personal consumption expenditures (PCE) are evaluated separately in the model. Bayesian vector auto regression (BVAR) The model has 7 variables namely PCE, PCE inflation, real GDP, federal fund rate, core PCE inflation among others where the model uses long time-series data to for quarterly inflation forecasts (Mukherjee & Kachwala, 2009). Bayesian methods with steady state (SS-BVAR) The model allows addition of information which is generated from outside the model. The additional information is vital during the forecasting of short term basis. The variables can include the financial variables or labor market variable. It takes into account time series for inflation dynamics (Mukherjee & Kachwala, 2009). Importance of improving the forecasts It promotes Stock Keeping Unit (SKU) for the satisfaction of demand. The forecasts necessitate automatic operations for constant analyst input without manual intervention. The methods necessitate the determination of the quantities that should be bought, manufactured and shipped. This is because the distribution operation processes involve the movement of raw materials to the finished goods for customers’ consumption. It helps the management to forecast on future demand based on the prevailing economic context. The management will be able to forecast future economic demand and supply based on the current economic situation (Mukherjee & Kachwala, 2009). The marketing management can use forecasting for long term and short term analysis so as to predict market share and market demand for planning purposes. The short term forecasts will also assist in setting sales quotas and targets. Recommendations/justification EBBD should qualitative and quantitative methods for the realization of its long term and short term objectives. Under the qualitative methods the company can use surveys to forecast on its incoming demand. The company can also employ consensus methods whereby they can use a specific customer group to make general forecasts. Under the quantitative method, test marketing can be used as a simulation before launching a new product line. The company can also forecast for the long term for its uncertain economic conditions. BVAR and SS-BVAR are the best inflation forecast methods since they use time series data. References Brandimarte, P., & Zotteri, G. (2007). Introduction to distribution logistics. Hoboken, N.J: Wiley-Interscience. Wang, J., & Wang, S. (2010). Business intelligence in economic forecasting: Technologies and techniques. Hershey, PA: Information Science Reference. Mukherjee, P. N., & Kachwala, T. T. (2009). Operations management and productivity techniques. New Delhi: PHI Learning. Read More