You are on page 1of 1

Linear regression

- Y is response variable, x is predictor


- Coefficients represent b0,b1
- Hypothesis test: check to see if there is
actually a relationship between quantitative
( variables in the population.
- R-squared: fraction of data’s variation
accounted for in the model
- 1-R-squared: fraction of original variation
left in the residuals.

Simulation
- 68,95, 99.7 Rule: in normal, bell shaped distribution, 68% of values fall within 1
standard deviation of the mean, 95% in 2, and 99.7% in 3 SDs.
- Monte Carlo Simulation: used to evaluate the impact of uncertainty on a
decision.
- Simulation results help us to make decision recommendations for the
controllable inputs that address not only the average output but also the
variability of the output. For example, an output can be net profit.
- Function for random # in excel = highest # of random value wanted * RAND.
- Example: 100*RAND()
- Random value = lower bound cell (A) + RAND()*(upper bound cell-lower)
- A is upper bound, B is lower bound
- Select both columns, click data, what if analysis, data table, leave the first cell
blank, then input a random cell for the second.
- From random values range, create a histogram, and regenerate the values by
clicking F9.
Normal Random Values: do the same as above, but using mean and standard deviation
= Norm.Inv(RAND(), mean cell, Std. dev. cell)
Discrete Distribution: With lower bound, upper bound, variable and profitability.
Random value = VLOOKUP(RAND(), whole chart, column, TRUE/FALSE). Same as
before to finish random value generation.

PROFIT SIMULATION
Start off by creating a table for each variable (include low bound, upper bound, cost profitability, and cumulative. (lower bound 1 always 0). Calculate up and
low bounds (same as above. Cumulative is upper bound, probability is given.
Parameters – selling p (given), other variables = VLOOKUP(RAND(), select upper and lower bounds, data column, TRUE/FALSE
Profit per unit = param 1 – param 2 – param 3 – param 4
Simulation trial, calculate what-if analysis
Output = MIN (profit range) MAX = MAX (profit range) Mean = average (profit range) Standatd dev. = STDEV.S (profit range)
Q1, Q2, Q3 = quartile. Inc.(profit range,1)*change numer for each P(profit<$5) = COUNTIF(profit range,”<5”)/number of trials
IQR – Q3 cell – Q1 cell *$5 changes with question)

You might also like