Problem

You need to compute confidence intervals for certain estimates and you're not sure what support Excel provides for these calculations.

Solution

Excel offers a built-in function called `CONFIDENCE` that allows you to compute confidence intervals based on a normal distribution. Moreover, Excel provides functions such as `NORMSINV` and `TINV` that allow you to compute inverses for normal and Student's t-distributions, respectively. You can then use these results to compute confidence intervals.

Discussion

In Recipe 5.1, I showed you how to use Excel's built-in functions to compute summary statistics. I also showed you how to compute summary statistics using the Descriptive Statistics tool available from the Analysis ToolPak. In both cases, confidence intervals for the mean were computed and I pointed out that the confidence interval returned by Excel's `CONFIDENCE` function is different from that returned by the Descriptive Statistics tool. Figure 5-1 shows the different values.

The difference in the computed confidence interval is due to the fact that the `CONFIDENCE` function uses the inverse of a standard *normal distribution* to compute the confidence interval, whereas the Analysis ToolPak uses the inverse of *Student's t-distribution*. Instead of using the `CONFIDENCE` function, you can use the functions `NORMSINV` or `TINV` to compute confidence intervals; this gives you the option of using a normal distribution or Student's t-distribution, which is more suitable for small samples.

`NORMSINV` computes the inverse of a standard normal distribution with a mean of 0 and standard deviation equal to 1. In other words, `NORMSINV` returns the *z-score* corresponding to a given probability. The syntax for `NORMSINV` is `=NORMSINV(`*probability*`)`. For a 95% confidence interval, you need to enter a probability of `1-0.05/2` or `0.975`. Referring to the example shown in Figure 5-1, you can compute the confidence interval for the mean using the formula `=H9*NORMSINV(0.975)`, where cell H9's value contains the standard error of the mean. This formula returns a confidence value of 0.015486, which agrees with the results returned using the `CONFIDENCE` function.

You can use `TINV` to compute the confidence interval based on Student's t-distribution. `TINV` computes the inverse of Student's t-distribution given a probability and the degrees of freedom characterizing the distribution. Again, referring back to the example in Figure 5-1, you can compute the confidence interval using the formula `=H9*TINV(0.05,H20-1)`. In this case, the probability for 95% confidence is entered as `1-0.95` or `0.05`. As before, H9 refers to the cell containing the standard error of the mean. Cell H20 contains the value returned by `COUNT`, and the number of degrees of freedom is computed by `h20-1`. The resulting confidence value is 0.015677, which agrees with the result returned by the Descriptive Statistics tool.

See Also

You can use the formulas discussed in this recipe to compute confidence intervals for other estimates as well. In Recipe 8.8, I show you how to compute confidence intervals for regression curves.

Using Excel

- Introduction
- Navigating the Interface
- Entering Data
- Setting Cell Data Types
- Selecting More Than a Single Cell
- Entering Formulas
- Exploring the R1C1 Cell Reference Style
- Referring to More Than a Single Cell
- Understanding Operator Precedence
- Using Exponents in Formulas
- Exploring Functions
- Formatting Your Spreadsheets
- Defining Custom Format Styles
- Leveraging Copy, Cut, Paste, and Paste Special
- Using Cell Names (Like Programming Variables)
- Validating Data
- Taking Advantage of Macros
- Adding Comments and Equation Notes
- Getting Help

Getting Acquainted with Visual Basic for Applications

- Introduction
- Navigating the VBA Editor
- Writing Functions and Subroutines
- Working with Data Types
- Defining Variables
- Defining Constants
- Using Arrays
- Commenting Code
- Spanning Long Statements over Multiple Lines
- Using Conditional Statements
- Using Loops
- Debugging VBA Code
- Exploring VBAs Built-in Functions
- Exploring Excel Objects
- Creating Your Own Objects in VBA
- VBA Help

Collecting and Cleaning Up Data

- Introduction
- Importing Data from Text Files
- Importing Data from Delimited Text Files
- Importing Data Using Drag-and-Drop
- Importing Data from Access Databases
- Importing Data from Web Pages
- Parsing Data
- Removing Weird Characters from Imported Text
- Converting Units
- Sorting Data
- Filtering Data
- Looking Up Values in Tables
- Retrieving Data from XML Files

Charting

- Introduction
- Creating Simple Charts
- Exploring Chart Styles
- Formatting Charts
- Customizing Chart Axes
- Setting Log or Semilog Scales
- Using Multiple Axes
- Changing the Type of an Existing Chart
- Combining Chart Types
- Building 3D Surface Plots
- Preparing Contour Plots
- Annotating Charts
- Saving Custom Chart Types
- Copying Charts to Word
- Recipe 4-14. Displaying Error Bars

Statistical Analysis

- Introduction
- Computing Summary Statistics
- Plotting Frequency Distributions
- Calculating Confidence Intervals
- Correlating Data
- Ranking and Percentiles
- Performing Statistical Tests
- Conducting ANOVA
- Generating Random Numbers
- Sampling Data

Time Series Analysis

- Introduction
- Plotting Time Series Data
- Adding Trendlines
- Computing Moving Averages
- Smoothing Data Using Weighted Averages
- Centering Data
- Detrending a Time Series
- Estimating Seasonal Indices
- Deseasonalization of a Time Series
- Forecasting
- Applying Discrete Fourier Transforms

Mathematical Functions

- Introduction
- Using Summation Functions
- Delving into Division
- Mastering Multiplication
- Exploring Exponential and Logarithmic Functions
- Using Trigonometry Functions
- Seeing Signs
- Getting to the Root of Things
- Rounding and Truncating Numbers
- Converting Between Number Systems
- Manipulating Matrices
- Building Support for Vectors
- Using Spreadsheet Functions in VBA Code
- Dealing with Complex Numbers

Curve Fitting and Regression

- Introduction
- Performing Linear Curve Fitting Using Excel Charts
- Constructing Your Own Linear Fit Using Spreadsheet Functions
- Using a Single Spreadsheet Function for Linear Curve Fitting
- Performing Multiple Linear Regression
- Generating Nonlinear Curve Fits Using Excel Charts
- Fitting Nonlinear Curves Using Solver
- Assessing Goodness of Fit
- Computing Confidence Intervals

Solving Equations

- Introduction
- Finding Roots Graphically
- Solving Nonlinear Equations Iteratively
- Automating Tedious Problems with VBA
- Solving Linear Systems
- Tackling Nonlinear Systems of Equations
- Using Classical Methods for Solving Equations

Numerical Integration and Differentiation

- Introduction
- Integrating a Definite Integral
- Implementing the Trapezoidal Rule in VBA
- Computing the Center of an Area Using Numerical Integration
- Calculating the Second Moment of an Area
- Dealing with Double Integrals
- Numerical Differentiation

Solving Ordinary Differential Equations

- Introduction
- Solving First-Order Initial Value Problems
- Applying the Runge-Kutta Method to Second-Order Initial Value Problems
- Tackling Coupled Equations
- Shooting Boundary Value Problems

Solving Partial Differential Equations

- Introduction
- Leveraging Excel to Directly Solve Finite Difference Equations
- Recruiting Solver to Iteratively Solve Finite Difference Equations
- Solving Initial Value Problems
- Using Excel to Help Solve Problems Formulated Using the Finite Element Method

Performing Optimization Analyses in Excel

- Introduction
- Using Excel for Traditional Linear Programming
- Exploring Resource Allocation Optimization Problems
- Getting More Realistic Results with Integer Constraints
- Tackling Troublesome Problems
- Optimizing Engineering Design Problems
- Understanding Solver Reports
- Programming a Genetic Algorithm for Optimization

Introduction to Financial Calculations

- Introduction
- Computing Present Value
- Calculating Future Value
- Figuring Out Required Rate of Return
- Doubling Your Money
- Determining Monthly Payments
- Considering Cash Flow Alternatives
- Achieving a Certain Future Value
- Assessing Net Present Worth
- Estimating Rate of Return
- Solving Inverse Problems
- Figuring a Break-Even Point

Index

Excel Scientific and Engineering Cookbook (Cookbooks (OReilly))

ISBN: 0596008791

EAN: 2147483647

EAN: 2147483647

Year: N/A

Pages: 206

Pages: 206

Authors: David M Bourg

Flylib.com © 2008-2020.

If you may any questions please contact us: flylib@qtcs.net

If you may any questions please contact us: flylib@qtcs.net