ICAP PRC-2 Quantitative Analysis for Business key terms

Interest calculated only on the original principal for the whole period, so the interest earned each year is identical and never itself earns interest.

Interest added to the principal at each compounding interval so that later interest is earned on earlier interest, producing growth that accelerates over time.

The true yearly growth rate after allowing for compounding within the year; it exceeds the nominal rate whenever compounding occurs more than once annually.

The amount that must be invested today at a given interest rate to produce a specified future sum, found by discounting that future sum backwards.

The multiplier, equal to one divided by one plus the periodic rate raised to the number of periods, that converts a future cash flow into its present value.

A series of equal cash payments made at regular intervals for a fixed number of periods, whose combined present or future value can be computed with standard formulas.

The sum of the present values of all expected cash inflows minus the present values of all outflows; a positive result suggests the project adds value.

The arithmetic average found by adding all observations and dividing by their count; it uses every value and is pulled strongly by extreme observations.

The middle value of a data set arranged in order; for an even count it is the average of the two central values and it resists distortion by outliers.

The observation or category that occurs most frequently; a data set may have none, one, or several modes and the mode works for qualitative as well as numerical data.

The square root of the variance, expressing typical distance of observations from the mean in the original units of the data.

The average of the squared deviations of observations from their mean; squaring removes sign and penalises large deviations heavily.

The standard deviation divided by the mean, usually shown as a percentage; it compares relative variability across data sets measured in different units or with very different averages.

One of three values that split ordered data into four equal parts, with the interquartile range between the first and third quartiles describing the spread of the middle half.

A measure of asymmetry in a distribution; positive skew has a long right tail pulling the mean above the median, while negative skew has a long left tail.

A chart of continuous data in which bars touch and bar area, not height alone, represents the frequency of each class interval.

A table grouping observations into class intervals and recording how many fall in each, forming the basis for charts and for grouped estimates of mean and spread.

A cumulative frequency curve plotted against upper class boundaries, used to read off medians, quartiles and other percentiles graphically.

Two or more events that cannot occur together in the same trial, so the probability of either one happening is the sum of their individual probabilities.

Events where the occurrence of one gives no information about the other, so the probability of both equals the product of their separate probabilities.

The probability that an event occurs given that another event is known to have occurred, calculated by dividing the joint probability by the probability of the known event.

A rule for revising a prior probability into a posterior probability after new evidence, combining the conditional likelihood of the evidence with the prior.

The distribution of the number of successes in a fixed number of independent trials, each with the same two possible outcomes and the same success probability.

A model for the count of rare events occurring in a fixed interval of time or space, defined by a single mean rate and often used as an approximation to the binomial.

A symmetric bell-shaped continuous distribution defined by its mean and standard deviation, with about 68 percent of values within one standard deviation of the mean.

A normally distributed value rescaled by subtracting the mean and dividing by the standard deviation, giving a mean of zero and standard deviation of one for table lookup.

The result that sample means from almost any population tend towards a normal distribution as sample size grows, underpinning confidence intervals and tests.

The probability-weighted average of all possible outcomes of a random variable, representing the long-run mean if the situation were repeated many times.

A complete listing, table or formula giving each possible value of a random variable together with its probability, whose probabilities must sum to one.

A diagram branching through decision points, chance events and payoffs, evaluated backwards to identify the alternative with the highest expected monetary value.

The payoff forgone by not choosing the best alternative for a given state of nature; expected opportunity loss is minimised by the same choice that maximises expected value.

A weighted price index that holds quantities fixed at base-period levels, so it tends to overstate price change because it ignores substitution towards cheaper goods.

A weighted price index using current-period quantities as weights, which tends to understate price change relative to Laspeyres and needs fresh quantity data each period.

The reference period against which later values are compared in an index, normally set to 100 so that changes are read directly as percentages.

Removing the effect of price change from a money series by dividing each value by a suitable price index, converting nominal figures into real terms.

A sequence of values of a variable recorded at successive equally spaced points in time, analysed as trend, seasonal, cyclical and irregular components.

A series of averages recomputed as the window slides along the data, smoothing short-term fluctuations so that trend and seasonal patterns become visible.

A figure expressing the typical percentage by which a season's value sits above or below the trend, used to deseasonalise data or to forecast within the year.

A forecasting method that updates the previous forecast by a fixed fraction of the latest forecast error, giving recent observations more weight than distant ones.

The proportion of total variation in the dependent variable explained by the regression line, equal to the square of the correlation coefficient in simple regression.

A number between minus one and plus one measuring the strength and direction of linear association between two variables, unaffected by which is treated as dependent.

The method of fitting a straight line by choosing slope and intercept that minimise the sum of squared vertical distances from observations to the line.

A strong statistical association between two variables that arises from coincidence or from a shared third factor rather than any genuine causal link.

A technique for maximising or minimising a linear objective subject to linear constraints, solved graphically in two variables or by algebraic methods such as the simplex algorithm.

The set of all points satisfying every constraint of a linear programme simultaneously; for two variables it is a polygon whose corners are the candidate optimum solutions.

The change in the optimal objective value from relaxing a binding constraint by one unit, showing the most a manager should pay for extra capacity.

A selection method in which every member of the population has an equal chance of inclusion and every possible sample of the chosen size is equally likely.

Dividing the population into internally similar groups and sampling randomly within each, which usually improves precision by ensuring every group is represented.

The probability distribution of a statistic such as the sample mean across all possible samples of a given size drawn from the population.

The standard deviation of a sampling distribution, showing how much a sample statistic is expected to vary from sample to sample and shrinking as sample size grows.

A range built around a sample estimate, together with a stated confidence level, that is expected to contain the true population parameter in that proportion of repeated samples.

The probability of rejecting a true null hypothesis that the analyst is willing to accept, chosen before the test and fixing the critical region.

The default claim of no effect or no difference that a statistical test attempts to contradict, retained unless sample evidence is sufficiently unlikely under it.

Rejecting a null hypothesis that is actually true, an error whose probability equals the chosen significance level of the test.

Failing to reject a null hypothesis that is actually false, an error whose probability falls as sample size or effect size increases.

The rule giving the probability that at least one of two events occurs as the sum of their probabilities minus the probability of both occurring together.

The rule giving the probability that two events both occur as the probability of the first multiplied by the conditional probability of the second given the first.