A guess about what is going to happen in the future. An integral part of almost all business enterprises. Logical and rational, but still a guess! Objective is to minimize error (as you will always be wrong!)
Lecture 1: The Forecasting Perspective What is a Forecast? • A guess about what is going to happen in the future • An integral part of almost all business enterprises • Logical and rational, but still a guess! • Objective is to minimize error (as you will always be wrong!) Why forecast? “It is far better to foresee even without certainty than not to foresee at all.” Henri Poincare, The Foundations of Science “The successful business executive is a forecaster first; purchasing, producing, marketing, pricing and organizing all follow.” Peter Bernstein, Against the Gods Uses of Forecasts Accounting Cost/profit estimates Finance Cash flow and funding Human Resources Hiring/recruiting/training Marketing Pricing, promotion, strategy MIS IT/IS systems, services Operations Schedules, MRP, workloads Product/service design New products and services Reputation may be made or lost • “I think there is a world market for maybe five computers.” (Chairman of IBM, 1943) • “Computers in the future may weigh no more than 1.5 tons.” (Popular Mechanics, 1949) • “There is no reason anyone would want a computer in their home.” (President, DEC, 1977) • “There are four ways economists can lose their reputation Gambling is the quickest, sex is the most pleasurable and drink the slowest But forecasting is the surest.” (Max Walsh, The Age, 1993) 27/08/2008 • “Trong thời gian tới, khả thị trường chứng khoán Việt Nam thấp ngang khó Mặc dù thị trường có điều chỉnh song thị trường chứng khoán tăng tăng mạnh cuối năm” – nhận định ông Nguyễn Thanh Tùng – chuyên gia tư vấn Cơng ty chứng khốn Artex • Đờng quan điểm đó, chun gia tư vấn – phòng phân tích cơng ty chứng khốn Artex, ơng Ngũn Văn Chiến khẳng định: “Hiện khó khăn thị trường tờn ngắn hạn dài hạn thị trường chứng khoán Việt nam tăng trưởng tốt”. History of Forecasting • The ancient Egyptians foretold harvests from the level reached by the Nile in the flood season • Non scientiffic forecasting: e.g Astrology, Book of Changes has a long history • Development of the quantitative methods: middle-to-second part of the 20th century Current Situation in Forecasting • Forecasting is widely used in business • There exist a well defined set of quantitative forecasting methods • There exists computer software that may be quite simply applied in forecasting • Excel program allows to solve simple forecasting tasks Forecasting Methods • Qualitative (subjective): people expertise – Little or no quantitative information available, but sufficient qualitative knowledge exists – Example (Delphi Method): • Individual opinions are compiled and reconsidered • Repeat until overall group consensus is reached • Quantitative (objective): math models – Sufficient quantitative information available – Example: predict the growth of revenue based on sales data of the previous 10 years Causal vs Time Series • Causal model: – Explain the relationship between the variable to be forecast and a number of independent variables – GNP = f(monetary and tax policies, inflation, capital spending, imports, exports) + Error • Time series model: – Relate the value of a variable(s) at one time point with values of the variable(s) at previous time points – GNPt+1 = f(GNPt, GNPt-1, GNPt-2, ) + Error Forecasting Methods Forecasting Techniques Qualitative Models Time Series Methods Delphi Method Naive Jury of Executive Opinion Moving Average Sales Force Composite Consumer Market Survey Weighted Moving Average Exponential Smoothing Trend Analysis Causal Methods Simple Regression Analysis Multiple Regression Analysis Seasonality Analysis Multiplicative Decomposition Jury of Executive Opinion Method • Appropriate managers within the organization assemble to discuss their opinions on what will happen to sales in the future The Sales Force Method • Predict future sales by analyzing the opinions of salespeople as a group • Salespeople continually interact with customers, and from this interaction they usually develop a skill for predicting future sales Consumer Market Survey • The method is useful for predicting the sales of a new product - there is no historic or past data available to forecast • Field surveys are conducted to gather information on the intentions of potential customers Delphi Method • STEP - Various Experts are asked to answer, independently and in writing, a series of questions about the future of sales or whatever other area is being forecasted • STEP - A summary of all the answers is then prepared No expert knows, how any other expert answered the questions • STEP - Copies of summary are given to the individual experts with the request that they modify their original answers if they think it necessary • STEP - Another summary is made of these modifications, and copies again are distributed to the experts This time expert opinions that deviate significantly from the norm must be justified in writing • STEP - A third summary is made of the opinions and justifications, and copies are once again distributed to the experts Justification in writing for all answers is now required • STEP - The forecast is generated from all of the opinions and justifications that arise from step Conditions for Time Series Forecasting • Information about the past is available • This information is in form of numerical data • The past pattern will likely continue in the future Elements of a Good Forecast Timely Reliable n i n a e M l u gf Accurate Written y s Ea to e s u “Laws” of Forecasting • Forecasts rarely perfect because of randomness • Forecasts more accurate for groups than individuals • Forecast accuracy decreases I see that you will as time horizon increases get a 10 this semester Forecasting Process Identify the purpose of forecast Collect historical data Plot data and identify patterns Check forecast accuracy with one or more measures Develop/compute forecast for period of historical data Select a forecast model that seems appropriate for data Is accuracy of forecast acceptable? No 8b Select new forecast model or adjust parameters of existing model Yes 8a Forecast over planning horizon Adjust forecast based on additional qualitative information and insight 10 Monitor results and measure forecast accuracy ... computer in their home.” (President, DEC, 19 77) • “There are four ways economists can lose their reputation Gambling is the quickest, sex is the most pleasurable and drink the slowest But forecasting. .. model: – Relate the value of a variable(s) at one time point with values of the variable(s) at previous time points – GNPt +1 = f(GNPt, GNPt -1 , GNPt-2, ) + Error Forecasting Methods Forecasting Techniques... of summary are given to the individual experts with the request that they modify their original answers if they think it necessary • STEP - Another summary is made of these modifications, and