All language subtitles for 05 - Forecast future results

- [Instructor] When you own or operate a business,

you're always wondering what happens next.

In this movie,

I will show you a couple of techniques

you can use to answer those questions,

at least as far as current trends allow.

My sample file is 02_05_Forecast,

and you can find it in the Chapter 02 folder

of the Exercise Files collection.

I'm on the Trend worksheet in this workbook,

and you can see that I have quarterly sales data

for the years of 2021 and 2022.

What I'd like to do is to get a forecast

based on those values for the quarters in 2023.

And you can do that using the fill handle.

If you're not familiar with the fill handle,

I will give you a quick demonstration of how it works.

In cell D2 I'll type a 1,

and then in cell D3 I'll type a 2.

Then I will select cells D2 and D3.

If you haven't used the fill handle before,

it is the green square at the bottom right corner

of a selected cell or cell range.

If you want to drag the fill handle to extend values,

select the two cells.

You can also do more as you'll see in a moment.

Move your mouse pointer over the fill handle,

it changes to a black crosshair.

And then drag down, and you see that it extends the series.

If I make the second value a 3

and then select the two cells and then drag again,

you see that it goes up, keeping the increment.

This is a way of extending the trend linearly,

so it's based on the line of best fit for this data.

I'll go ahead and delete the values that I have selected

just so they're not cluttering up the worksheet.

And then I can extend the values in B2 through B9

by selecting those cells.

And then I will grab the fill handle and drag down.

And you can see that based on the previous values,

if we have a linear trend in our data

we'll have 412,000, 426,000, 439 and 453.

So even though there were increases

and decreases based on quarters,

overall the trend is up.

So this analysis is useful in that it tells you

that your sales are generally trending positively.

You're making more money per quarter.

However, it doesn't account for any increases or decreases.

For example, do you sell more in quarter three

or less in quarter four?

So all this does is give you a general idea.

Now I'll switch to the forecast worksheet,

and I will show you how to use a formula

to create a different kind of forecast.

So click the Forecast sheet tab,

and I can see here that I have data about the distance

that an individual traveled to get to a store

and then the amount that they spent once they got there.

And I can scroll down,

and I see that I have about 26 rows of data.

Let's say that I'm interested in predicting

how much individuals who live 30 miles from the store

would spend if they came.

The scenario I have in mind

is that we might target online ads to potential customers

who live in a neighborhood about 30 miles away.

So I will go to cell D2 and type 30.

And then in cell E2, I will type my forecast formula.

So I'll start by typing forecast,

and then I will use FORECAST.LINEAR.

So I'll move down to highlight that because I don't,

in this case, want to use any sort of exponential smoothing,

which is an advanced technique

and is beyond the scope of this course.

So I have linear forecast, press Tab.

And then the X is in cell D2,

that's the number of miles driven, then a comma,

the known Ys.

And it is easy to reverse the order of these two columns.

The known Ys are the amount that were spent,

so this is the dependent variable,

and it is dependent on the distance driven.

So I'll click cell B2, then Control + Shift + down arrow.

And I've selected B2 through B27, scroll back up,

and then type a comma.

And the known Xs are in A2 through A27,

and I can just type that in.

Right parenthesis and Enter.

And I see that the predicted amount spent is about $56.29

if they came from 30 miles away.

And looking at the data around that point,

I see that I have people who traveled 50 miles,

and they spent 67, 125, and 75.

Okay, a little bit more.

45 spent 35, 22 40,

20 25, and 20 50.

So the predicted amount of 56.29, makes sense,

giving the data that surrounds that point.

One thing you shouldn't do is to use your data

to make a prediction or forecast

for a value that is outside of your data range.

So if I click cell D2 and type in 200 and Enter,

you see that I get a predicted amount spent of 248.16.

And that is a valid amount.

The problem of course, is that my data only goes up,

as we see here in cell A27, to 152 miles.

And the question is whether someone

who drives from a long way away

would spend that much money.

For example, if I type in 300 and Enter,

we get a predicted amount of 361,

which is almost twice as much as the amount

that the individual who came in from 152 miles away spent.

You can use the FORECAST function

to predict amount spent, in this case,

for miles driven values

that are a little bit outside of the collected data range.

But in most cases, you will be safest

if you stay well within the parameters of the data

that you collected earlier.