Advanced Excel/ Excel based analysis and
modelling
Introduction
Excel is commonly known for creation and manipulation of tables of data and
presentation of data in diagrams. Excel can do so much more which include
-Automatize the manipulation of tables through modification of existing routines and
creation of new applications
-Data analysis, modelling and simulation
Advanced Excel skills are very applicable in Industry, Finance and Engineering.
Some basics(refreshing)
-enter data
-modify entered data
-format cell entries
-format cell size
-merge cells
-add comments to cell(s)
-formulae
-autofill function
-cell reference (relative, mixed and absolute)
Excel assumes the reference is a relative reference, that means
the cell reference changes when you copy the cell.
A column or a row can be “fixed” by adding a “$”-symbol:
There are four possibilities:
= A3  changeable column and row (relative reference)
= A$3  changeable column, fixed row (mixed reference)
= $A3  fixed column, changeable row (mixed reference)
= $A$3  fixed column and row (absolute reference)
-naming cells: names can be used as variables in a formula instead of lengthy
references
LiSiz 2022 @Stats UZ
-Excel functions-built in: syntax =function_name(argument(s))-arguments are
separated by commas if more than argument is required for the function. A function
processes input and returns a value. When you select a function from the list, you are
presented with a dialog box/window which allows you to supply input/argument
values for the function. You can supply the arguments directly by typing in or use the
toggle tabs to access the worksheet for point and selecting cell ranges (left mouse
click) and back to the dialog box.
Common Excel error return values
#DIV/0! ; results from dividing by 0
#NAME?: a formula contains an undefined variable or function name, or the syntax is
not valid
#N/A: formulae refer to cells which do not contain the appropriate data
#NULL!  a result has no value
#NUM!  value not an element of the real set, eg square root of a negative number
#VALUE!  invalid argument type
#REF!  invalid cell reference
circular error  a formula contains a reference to its own location
Conditional functions
IF (upto 7 nestings)
Implicit If functions: countif, sumif
Syntax = SUMIF(range,condition,sum_range)
LOOK UP functions
=VLOOKUP(lookup_value, table_array, column_index,match)
=HLOOKUP(lookup_value, table_array, row_index,match)
table_array  The range reference or name of the lookup table.
column(row)_index ; The column (row) of the table from which the value is to be
returned.
LiSiz 2022 @Stats UZ
match  Is a logical value, i.e. TRUE or FALSE, which specifies whether you want an
exact or approximate value. It is optional with default value TRUE.
Worksheet/Workbook Protection
It is possible to protect a workbook, a worksheet or parts of a worksheet. (Under
Review menu)
You can lock or hide formulas of parts of a worksheet by highlighting the cell range
then right mouse click, select Protection. Then after you still need to lock the
protection.
EXCEL Statistical utilities
Presentation of Quantitative data
Data classification, Types of charts and graphs, pivot tables
Analysis of quantitative data
The universal goal of Data analysis is to answer one very important question—
what does the data reveal about the underlying system or process from which
the data is collected?
Excel provides you with numerous internal tools designed explicitly for data
analysis, the user is also capable of creating their own unique creation to perform
many types of analytical procedures by
using Excel’s basic mathematical functions. This is often how an add-in is born:
an individual creates a clever analytical application and makes it available to
others.
An add-in is a program designed to work within the framework of Excel. Add-ins
use the basic capabilities of Excel (for example, either Visual Basic for
Applications (VBA) or Visual Basic (VB) programming languages) to perform
internal Excel tasks. These programming tools are used to automate and expand
Excel’s reach into areas that are not readily available. There are many free and
commercially available statistical, business, and engineering add-ins that provide
capability in user-friendly formats.
(1)Excel provides resident add-in utilities that are extremely useful in basic
statistical analysis. The Data ribbon contains an Analysis group with almost 20
statistical Data Analysis tools.
(2) Excel provides dozens of statistical functions through the function utility ( fx
Insert Function) in the Formulas ribbon, Select the Statistical category of
functions in the Function Library group, select the function you desire, and insert
the function in a cell. The statistical category contains almost 100 functions that
relate to important theoretical data distributions and statistical analysis tools.
(3) There are numerous commercially available add-ins—functional programs
that can be loaded into Excel that permit many forms of sophisticated analysis.
LiSiz 2022 @Stats UZ
Presentation of Qualitative Data—Data Visualization and analysis
Sorting and filtering, Pivot tables
Inferential statistics
Chi-square tests, confidence interval, ANOVA, Experimental designs
Visual Basic for Applications (VBA)1
Excel has a powerful programming language which you can use to write your own
programmes/functions: thereby automating tasks. This offers a solution for handling
repeating operations more efficiently.
VBA code is entered in the Visual Basic Editor (VBE). VBA is installed by default,
accessing the Visual Basic Editor (VBE), through which you'll enter your VBA code,
doesn't automatically appear in the Ribbon.
To activate Click on File, then Options, then Customize Ribbon, then Developer: The
Developer Menu will be added and you can access the VBE tab. Alternatively you can
use ALT+ F11 to access VBE
Creating a used defined function
First activate the Visual Basic Editor (VBE). The VBE is like most other applications.
It is equipped with a menu and a toolbar at the top of the window
and has several subwindows:
- The Project Explorer displays the hierarchical structure of projects.
- The Properties Window displays the properties of the projects.
- The Module Window contains the VBA-code of your project.
- The Immediate Window displays compiling messages.
- The Module Window might not be visible when you open VBE.
VBE menu bar: Insert
Module (LC) (LC=Left Click on the mouse)
- The Immediate Window is made visible by
VBE menu bar: View
Immediate Window (LC)
Steps for writing a program
Writing any kind of computing program consists of three basic principal steps:
i) Design an algorithm which will perform the task you want.
ii) Translate the algorithm into a computer language (code) with a certain syntax, e.g.
VBA in our case.
iii) Test (debug) your program thoroughly
1
Adapted from Andreas
LiSiz 2022 @Stats UZ
Fring’s handout
Program structures types
(i)
sequential structures (line by line)
1 ...................
2 ...................
3 ...................
(ii)
control structures
· branching or decision structures
(iii)
looping (repetition structures)
(iv)
controlled GOTO
NOTE: It is useful to draw flow charts in order to keep track of the logic of the
program structure. You do not need to write all comments in detail, but it suffices to
write general statements in words.
LiSiz 2022 @Stats UZ
General structure of a sequential User Defined Function
Function name [(arguments) [As type] ] [As type]
[statements]
[name = expression]
[Exit Function]
[statements]
[name = expression]
End Function
- name  the name of the function
- arguments  a list of input value (just like for built-in functions)
- type  the data type which will be returned by the function
- statements  valid VBA commands
- expression  an arithmetic expression assigned to the function name, which will be returned
· Everything in bold has to be typed exactly as above.
· Everything in squared brackets [...] is optional.
· Each statement has to begin in a new line.
· In case the statement is longer than the line you can split it by typing “ _” (i.e. space
and underscore).
· A program (function) is read from top to bottom, that is each line is executed after
the next. There might be branches, loops etc which you can design.
· When End Function or Exit Function is reached the calculation terminates and the
value last assigned to the function´s name is returned.
- An assignment is done by an equation, which has to be read from the right to the left,
i.e. the value on the right hand side of the equation is assigned to the name on the left
hand side · The arguments are the Input and the function name contains the Output.
Examples
(i)Function F(x)
F = 3 * x + 10
End Function
You can now use this function on an Excel worksheet in the same way as you use a
built-in function, e.g. “=F(5)” would return 25
(ii) Function FF(x)
b=2*x
FF = b + 5
End Function
The variable b only exists temporarily inside the function FF.
LiSiz 2022 @Stats UZ
(iii) Function G(x,y,z)
G = y*x + z
End Function
As for built-in functions you can have more than one input variable (argument).
(iv) Function Q(a,b,c,x)
' quadratic equation
Q = a*x^2 + b*x +c
End Function
You can add comments to enhance the readability. VBA does not execute text
following a single quote. “=Q(2,4,5,3)” returns 35 ( 2*3^2+4*3+5)
(v) Function S(x, y, z)
S = 2 * Application.WorksheetFunction. FunctionName (x, y, z)
End Function
You can use Excel built-in functions inside user defined functions e.g. FunctionName
= SUM. “=S(1,2,3)” will return 12
Naming your User defined functions
- The first character in the name has to be a letter.
- The names are not case sensitive.
- Names are not allowed to contain spaces, @, $, #,... or be identical to VBA
commands.
Errors on debugging
- Inevitably you will make some mistakes either just typos or
structural ones and you need some strategy to eliminate them.
- Some mistakes block the entire WS, e.g. suppose you type:
Function Err(x)
Err = 2 * Sqr
(Here the brackets are missing in Sqr)
End Function
- Call this function on the WS (Recalculation of the WS is F9) and an error message
will be displayed : Left mouse click on OK and the mistake will be highlighted then
Unlock with “Reset”
LiSiz 2022 @Stats UZ
Declaring variable types
- Recall: Function name [(arguments) [As type] ] [As type]
- The first type refers to the variable type of the arguments and the second type to the
variable type of the function.
- You can also declare variables used inside the program:
· Syntax: Dim variable_name as type
· Declaring the type avoids that different types of data get mixed up. You can trace
systematically mistakes in long programs.
- When you do not declare the type it will be “variant” by default.
· The variant type takes more space than properly defined variables. Your program
will run faster when you declare the types
Variable types
· integer  integer numbers 0, ± 1, ± 2, ± 3, ... ±32767, -32768
e.g. Dim a as integer
a = 32768 gives an error
a = 11.3 implies a = 11
· boolean : 16 bit (2 byte) number which is “true“ or “false“
· string  can contain up to 2 billion (2^31) characters
· single  32 bit (4 byte) floating point number between
-3.402823E38 to - 1.401298E-45 and
1.401298E-45 to 3.402823E38
· double  64 bit (8 byte) floating point number between
-1.79769313486231E308 to -4.94065645841247E-324
4.94065645841247E-324 to 1.79769313486232E308
Expl.: Dim a as integer
a = 32768 Ø gives an error
a = 11.3 Ø a = 11
· variant  16 byte with numerical value (here you see the disadvantage, consumes
more disk space)
· date  64-bit (8-byte) number representing
dates from 1-st January 1900 to 31-st December 9999
and times from 0:00:00 to 23:59:59.
Working with dates and times
VBA handles dates as numbers where
1-th of January 1900 = 1
2-nd of January 1900 = 2
........
25-th of October 2005 = 38650
- Some Date and time related VBA-functions:
· Month(date) : a number between 1 and 12 representing the month
· Weekday(date) : a number between 1 and 7 representing the day
LiSiz 2022 @Stats UZ
· Year(date) : a number between 1900 and 9999 for the year
· Hour(date) : a number between 0 and 23 for the hour
· Minute(date) : a number between 0 and 59 for the minute
· Second(date) : a number between 0 and 59 for the second
Examples:
a) Write a user defined function which computes the weekday for a date
Function DD(da As Date)
DD = Weekday(da)
End Function
· Format the cell A1 as date and enter 25/10/2005
· “=DD(A1)” returns 3
b) Write a UDF which calculates the age in years given the birthdate.
Function age(birthdate As Date)
age = Int((Now() - birthdate) / 365)
End Function
· (Now() - birthdate)  the age in days
· Int( x )  extracts the integer part of x
· age  the age in integer numbers of years
LiSiz 2022 @Stats UZ