Forecasting 630 - Notes Finance 630 is a cocktail of: Assumption(s): wrong/right, leading to wrong/right results Finance Macro Economics Statistics Assumption – something already taken as guaranteed Finance – normal, rational, applied behavior of humans Macro economics – everyday approach to economics Statistics – numerical effects or facts from figures Forecasting – predictions, forward looking Precise – highly definite Predictions of outcomes are rarely precise Philosophy of a CFO: C = crisis F = fluctuation O = optimism U = uncertainty Industry – refers to production of goods and services Commerce – refers to distribution. When industry finishes job, commercialization begins. Regular business. Any word in business except “production.” Uptown – residential Downtown – business Trade – branch of commerce. Many trades make one commerce Business – summation (∑) of industry, commerce, trade. Micro – (i) individual item. 1 single item. Macro – (a) all/aggregate. ∑ of all micros. Business -> macro -> industry -> micro Production – quantity. Number of goods or services you produced. U.S. is currently #1 in production Productivity – Quality. = cost of production (COP). Japan is #1 Germany – land of innovation U.S. – land of articulation Japan – land of imitation Always ask why; leads to culprit. Chapter 1 Purpose is to make judgment for business in the face of uncertainty. Humans tend to be optimistic, underestimate uncertainty. Best way to forecast are quantitative techniques and good judgement. Steps of Forecasting Process: 1. 2. 3. 4. 5. Problem formulation and data collection Data manipulation and cleaning Model building and evaluation Model implementation (actual forecast) Forecast evaluation Human intervention is very important in forecasting Chapter 2 Population – all items of my interest Sample – subgroup of items of interest Sampling – segmenting by interest Mean () – all of values of population, divided by number in population Median (Md) – ½ above, ½ below. Mode (Mo) – a value happening most frequently. Standard deviation () – how far, or how you’re deviating from the mean. SD = +3 -> means you are 3 points above mean Range – difference between highest and lowest/max & min Smaller ranges are good. Ex: 52pts – 92pts. R = 40 Quartile = ¼. Divide data into four groups Cross section data – same time. Today. Same day. Price of milk across multiple stores on same day. Time series data – over time. Look at price of milk everyday over month. You can look at products, just means over time. Dot plot – data in points. Histogram – data in boxes Scatter diagram – visualize between two variables. For showing low/medium/high achievements quickly. Random variable – looking for one outcome by chance Random – outcome by chance. No math, no science. Lotto ball numbers Discrete random variable – specified value/figure/number (aka – discrete variable, or just discrete) Continuous random variable – within same range Normal distribution – nothing unusual Sampling distribution – something small that represents something larger. When micro represents macro. Sampling can be very wrong, but must start somewhere. Estimation – sample gives you inference, it is called estimation. Aka – forecasting Forecasting – estimation Estimation in statistics – forecasting in finance, or in business subjects. Ideas change from wisdom out of experience. Hypothesis – assumption(s) that make me happy Hypothesis testing – get an assumption and test it. Can be wrong, or accept it. Hypothesis testing: 1. Formulate the hypothesis being tested (null hypothesis) and state the alternative hypothesis. 2. Collect a random sample of items from the population, measure them, and compute the appropriate sample test statistic. 3. Assume the null hypothesis is true, and determine the sampling distribution of the test statistic. 4. Compute the probability that a value of the sample statistic at least as large as the one observed could have been drawn from this sampling distribution. 5. If this probability is high, do not reject the null hypothesis; if this probability is low, the null hypothesis is discredited and can be rejected with small chance of error. Hypothesis = 1 Hypotheses = multiple Can give you confidence either way, but does not provide forecasting. Null hypothesis – (H0) – hypothesis being tested. Can be accepted or rejected. H0 - Original hypothesis. After rejection name it – H1 H1 means alternative hypothesis Type I error – () – alpha - true Ho is rejected. Truth denied. First error in hypothesis testing. Type II error – () - beta - false Ho is accepted. Second error in hypothesis testing Correlation – how two elements relate. Relationships. Never questions causation. Price up, quantity sold down = negative correlation Price up, quantity sold down = positive correlation Regression analysis – we estimate the equation that best fits sets of observations of dependent variable and independent variables. E(x) – expected value of x P(x) – probability of x - (x bar) sample mean df – degrees of freedom Population parameter – population quantity S – sample standard deviation r – correlation coefficient for sample data The expected value of a random variable is the mean value of the variable over many trials or observations. The binomial distribution is a discrete probability distribution describing the likelihood of X successes in n independent trials of binomial experiment. The Z-score of any X value is the number of standard deviations from the central value of the curve () to that value. The normal distribution has a bell shape and is completely determined by its mean and standard deviation. A point estimate is a single-valued estimate of a population parameter. An interval estimate is an interval within which the population parameter is likely to lie. One-sided/one-tailed test – the alternative hypothesis specifies values of the population parameter to one side of the value specified in the null hypothesis. Two-sided/two-tailed test – values of the test statistic much larger or much smaller than the value of the population parameter under H0 lead to rejection. The p-value or significance probability is the probability of getting at least as extreme a sample result as the one actually observed if the H0 is true. Equivalently, the p-value may be regarded as the smallest for which the observed test statistic lead to the rejection of H0. Positive linear relationship – as X increases, so does Y Perfect positive linear relationship – as X increases, Y increases also, and in a perfectly predictable way. The X and Y data points appear to lie on a straight line. Imperfect positive & negative linear relationship – as X increases in scatter diagrams, Y increases or decreases but not in a perfectly predictable way. The X-Y points do not lie on a straight line. Linear relationships – the X- relationship, be it perfect or imperfect, can be summarized by a straight line. Slope of the line – the slope of any straight line is defined as the change in Y associate with a one-unit increase in X Correlation coefficient – measures the extent to which two variables are linearly related to each other. Chapter 3 First thing you must do is find pattern in data. Data pattern tells you what technique to use. GIGO – garbage in, garbage out If data is wrong, no honest forecasting. Data is used to find validity, reliability, and accuracy. Data should be: 1. 2. 3. 4. Reliable and accurate Relevant Consistent Timely Follow a system or technique for results Time series data – collected over time period Cross sectional data – over single time single point Horizaontal data – fluctuate around a constant level. Virtually constant more or less. (time series) Trend – growth or decline in the time series over an extended period of time. (time series) Cyclical – wavelike, goes up and down Seasonal data – repeats itself year after year Choosing forecast technique: 1. Why needed? 2. Who will use? 3. Continued on pg 77 Chapter 4 Chapter 5 Time is the glue of economic progress Time series – observations of a variable that become available over time Decomposition – separate into parts under some position Index numbers – percentages that show changes over time Business indicators – business related time series that are used to help assess the general state of the economy. (East Germany – time series proof that communism is not a good theory) Chapter 6 Regression line tells you two things: 1. Two variables, 1 independent & then 1 dependant 2. Dispersion Standard error of the estimate – Variance – standard deviation squared ANOVA – analysis of variance Coefficient of determination – extent of relationship Analysis of residuals – 239 Residual – difference between forecast result and actual Growth curves – 246 4 things needed for good writing 1. 2. 3. 4. Theme Logic Language Consistence Imperfection – they know less than I know