シャドーイング練習: Forecasting METHODS - Qualitative and Quantitative Walkthrough in Excel - 動画で英語スピーキングを学ぶ
レッスンを作成中...
1
So there are several ways companies approach forecasting, especially today.
2
There are two methods that are usually used when it comes to forecasting a quantitative method or a qualitative method.
3
Your quantitative methods are your forecasting techniques that include time series.
4
Your moving average is your weighted moving average, your linear regressions and things of that nature.
5
And then you have your qualitative methods that are a little bit unique.
6
They're more so based on opinion or judgment of consumers and experts.
7
In most cases, these are usually most beneficial when you don't have any quantitative data.
8
So qualitative data is mostly used, not just it's used in specific industries, but it's also used when you have new products or industries like the real estate industry, because it changes year over year.
9
So that past data may not be as relevant today as it was yesterday.
10
Therefore, you have quality of approaches so that you don't always have to rely on those quantitative approaches for that may not be as effective in specific industries.
11
So let's talk about the qualitative methods first.
12
First, we have what we called a build up method.
13
Now this is where we basically start at the bottom of an organization and solicit information all the way up the chain.
14
Now, that can vary based on the organization.
15
You can start in a warehouse with a blue collar worker and work your way all the way up to the senior exit and all of the middle management and leaders in between to determine what a forecast for a particular product may be.
16
You can also start middle management and then work your way up to two or three levels to determine what a forecast the idea is to solicit information from different aspects of the organization to determine what the forecast should be.
17
The next method is survey method, and this is simply a survey, right?
18
We send out questions and do phone interviews or even face to face interviews to determine what a forecast should be.
19
This is where we get feedback.
20
Sometimes you could use the surveys, especially if you were writing a book.
21
You want to publish a new book and you wanted to know the level of interest of, let's say you were you're writing a romance novel.
22
Yeah. Or deep romance novel.
23
And you wanted to know people in a particular age group or market or region, their level of interest in romance novels.
24
And you can create a survey or conduct some type of phone interview to see that level of interest before you even write the book, right?
25
It's a form of qualitative forecasting because you have no quantitative data.
26
This is your first romance novel, so you could use a survey method to do something like that.
27
Next, we have a test market right?
28
A test market is when you take your product, you've produced your product, but before you mass produce your product, you use a test market to determine how people respond to it.
29
I believe something like this was done with the ring video doorbell.
30
The camera is really a security feature, but initially it was marketed as a doorbell, so you determine what price point people will pay for it.
31
You give it, give them the product free.
32
Maybe test it out, let us know what your thoughts are.
33
And then when you see that in this small region of people, everyone fell in love with the product that lets you know that you have a great product on your hands and you can mass produce this and expect the product to sell.
34
So a test market is a great way to go When you have a product that you know, consumers would rather try out first because of how unique or innovative that product is.
35
And another technique we have is what we call the panel of experts, also called the Delphi technique.
36
And this is where you solicit information from people that are knowledgeable about a particular subject to determine the forecast.
37
So in this case, you don't go to consumers and put out surveys for your romance novel.
38
You go to some experts that are maybe written some romance novels and know a little about what people are, what's in demand when it comes to a particular type of book, or what is it that people look for in romance novel novels?
39
You go to an expert or group of experts, for that matter, and you get that information from them instead of getting it directly from the consumer.
40
This is just another approach.
41
You could actually do both the panel of experts and the survey method to get this information when writing your new romance novel.
42
Right. Good luck with that anyway. Sometimes.
43
Or there is another opinion I think is called executive opinion, very similar to panel of experts, but you guessed it, it's just the executive's opinion.
44
And in organization they typically have a certain level of expertise when it comes to the products within that organization.
45
It can sometimes determine what a forecast should be.
46
So those are your qualitative methods of forecasting. Now let's talk about the quantitative methods of forecasting.
47
Now, at this stage in the game, there are several, several forecasting models that are used in all of these software programs that can get very complicated.
48
Not going to talk about all of that.
49
I'm going to talk about time series methods, and that is going to be your simple moving average or what we call your simple moving average and your weighted moving average.
50
Now, there are a few others that I'll mentioned exponential smoothing, linear regression.
51
Those are also forecasting methods that organizations quite use or people use in Excel all the time.
52
There are some others that I will not mention because I can't think of them.
53
But this is not all of the forecasting models or methods that are out there.
54
These are just a few that I'm discussing in this course.
55
So when it comes to moving averages, the reason moving averages is weighted and simple and weighted moving averages so important is because you can do a lot with these very simple calculations.
56
They can tell a lot about your product and give you accurate forecast predictions just based on simple averages.
57
So simple moving average is just that every time period is treated equally the same.
58
You take all of your products for a specific time period, add those up, and for that next month you take the average that is your forecast and you consistently do that moving forward.
59
Now, the time periods you choose are up to you.
60
Most companies choose based on their product or their industry.
61
They can use to last three months, last six, last year, last nine, last two years.
62
That is totally up to them.
63
But you sum all of the products sold over that period of time and that gives you your forecast for that next month of how much you should produce for the next time period.
64
And we'll look at an example in Excel to make this make a little bit more sense for you.
65
And then we have the weighted moving average.
66
So when it comes to the weighted moving average, it's a little bit different every time period.
67
It's not treated the same.
68
So we apply different weights, levels of emphasis to different time periods to determine how important that time period is for that month that you're forecasting for.
69
This is also called your seasonal index.
70
And it's really just comparing how specific seasons compare to the averages over that span of the time period that you selected.
71
And basically what I'm saying is this is where seasonality comes into play.
72
You heard me mention earlier about Valentine's Day.
73
Saint Patrick's Day.
74
So when you have different parts of the year, that product sale much more, much more during a specific ocular time or much less during a particular time.
75
This is when you want to use a weighted moving average and you can apply emphasis or more weight during those times where that product will sale in which you will produce more because people are going to buy more and then where they're not buying as much, you apply less emphasis and your weights are lower, right?
76
So that you don't produce as much during those times where the product does not sell.
77
Okay. So what I have here in this spreadsheet is just a simple forecast based on previous sales.
78
So on the left I have sales history and on the right I have the sales forecast.
79
Now this, as you can see, is in units.
80
Most companies will probably have this in dollars as well, so they can apply a monetary value to this or see a monetary value which would just include the price of these products here on the left.
81
And then a multiplication by the number of units produced in that to let them know how much money they made historically and how much they are projected to make in the future.
82
So going back to the simple moving average again, that is just when you take your number of time periods.
83
So if we take 12 months here, we see that over a 12 month period we so three, seven, four, four, right.
84
And that is divide it by 12 time periods and that gives us an average of 312 and that is our forecast for January 2022.
85
Now for February we would do the same thing, but we would go back 12 months, which would exclude January 2021, and we would include this January 2022, and that would give us that average of 328.
86
And that is basically what this bridge sheet is doing as it moves forward.
87
It moves back 12 months to generate a simple moving average for the next month's forecast, and it's doing that for each product.
88
So that's how a simple moving average works is simply taking the number of time periods and dividing that total amount by the number of time periods.
89
We could also do a three month average, right?
90
This total is 522, so the average is 174.
91
That means our forecast for January would be 174.
92
Or we could look at it from a six month perspective.
93
Right? That so is 2522 with an average of 420.
94
Our January 22 forecast would be 420 units.
95
Now let's look at a weighted moving average, which is can be a little bit more challenging here for this game or X series.
96
You can see the numbers are pretty sporadic, especially when we get to July 20, 21.
97
We go from selling roughly four or 500 a month to 1600, and that is where you can have problems using a simple moving average.
98
Needless to say, the same thing happens on a even more drastic scale at the end of 2021, where we jump to 2520 200 a month.
99
So if we're using a 12 month simple moving average, you can see that's going to generate an average of 910 for our forecast for January 22.
100
But we know on January 21 we only sold 100 units.
101
That's probably not going to be the most accurate.
102
So applying a simple moving average to this particular product is not going to be ideal because you can see that early in the year we're bringing in a ton of inventory, right?
103
We don't see that spike until around July and then the end of the month.
104
So we don't want to bring in this much inventory that early and that's where the weighted moving average comes into play.
105
So what I have here is I've broken it down by quarter and I basically take in the sum of what we saw for each quarter.
106
So that 1966 is quarter one and then that 1455 is quarter two, so on and so forth, recorded three and quarter four.
107
And then here I have the plot weight, right?
108
As you can see, that equals 100% or 1.0, 85% of that weight is going towards Q one, because that is the time period where we want to place the most emphasis.
109
And then for.
110
Q two, three and four, where we so more units, we're really only putting 5% towards each each of those quarters because we don't really want to factor that too much into our forecast because we know that's where a lot of the spikes took place.
111
So when we look at quarter one from a simple moving average perspective and we just look at these three months, we see we have a sum of 2905, right?
112
That is much more than the than 966 that we sold.
113
So there's three times it's more we don't want to bring in that much inventory.
114
So if we apply the simple move, I mean, the weighted moving average with these weights that I've selected, now when you're talking about doing this profession only using software systems and forecasting models, there would be some logical reasoning behind the numbers.
115
Select it for the weights, but in this case our randomly selected this.
116
So what what we see is when these weights are applied, we get these numbers.
117
We're basically multiplying the weight by the number so for each quarter and then we're going to add the numbers together and that's going to give us our forecast for Q1 2022.
118
So for Q1 2022, instead of producing the 2905 units, we're only going to produce 1318 and that is much more reasonable compared to deny 66 then the 2905.
119
And that is when a weighted moving average can be very beneficial to organizations because of those time periods where you need to apply more emphasis to a specific time period or less emphasis in a particular time period.
120
So if you recall earlier in the video I mentioned that forecasting can be used in many different aspects.
121
So what I have here is another example of how a weighted moving average can kind of put things in a to a different perspective.
122
So consider that.
123
Let's say you were looking to move to a new location.
124
You had four job offers and you had four job offers for different locations and you wanted to choose the job offer based on the factors that you see crime rate, cost of living, the school district and taxes.
125
And based on those factors, you want to make a determination of where you should move so that you could move to the best location for you and your family.
126
So what I have here is based on these factors, you've rated each factor for every city on a scale of 1 to 100, and the city with the highest number ideally is going to be the place that you should move to based on your factors.
127
So when we look at this first set of data, we see that London would come out on top with a total of 219, followed by Dallas.
128
And you can rank those right from 1 to 3 and four so that you can kind of get an idea of best to worst case.
129
Now, when we look at this, let's say we want to apply weights to this because there's a certain level of importance that these factors mean to you, meaning the cost of living and the crime rate is more important than the school district and the taxes.
130
So because of that, you go through and you apply different weights to each of these factors.
131
Again, those weights equal equals 100% or the sum of 1.0.
132
So now that we have these weights into play and we use the same numbers, we're basically taking our initial numbers and we're multiplying by the weight and that's giving us a new data set or a new set of numbers.
133
And then when we see those numbers up, as you can see, the rankings change a bit.
134
London is now number three and was previously number one and Dallas is now number one more so because the cost of living as well as the crime rate is where Dallas really has good numbers as as well as the school district.
135
So those top three really drives the needle specifically for Dallas.
136
But you see Dallas comes in first and then s followed by New York, then New London, followed by Atlanta.
137
But what I'm trying to portray here is when you look at this from a weighted moving average perspective, this could help you make a more sound decision based on your preference and not just based on the factors that are considered, but based on how those factors relate to you as an individual.
138
So that's it for forecasting methods.
139
Key takeaways you have to forecast the methods qualitative or quantitative.
140
Quantitative methods are usually used when you have past data that is accurate for your industry.
141
Again, not for real estate or sometimes even the car industry.
142
If interest rates fluctuate too much and things like that, they can really throw those industries off.
143
So last year's historical data is not not always an accurate depiction of what forecasting would look like for the next year.
144
However, companies that sell the same products over and over and over and over, that quantitative historical data is the way to go.
145
Now you're qualitative methods, your build up method, your survey, your test market, your Delphi technique or panel of experts, your expert opinion, all of those methods are used primarily because, again, a specific industry or you don't have any historical data.
146
Therefore you have to determine how you can forecast that romance novel, right?
147
So that is it.
148
Look forward to speaking with you in the next video.
📺 同じチャンネル
B21
What is Capacity Planning in Operations Management
Business EnglishUniversity StudentsWorking Adults
B21
What is Operations Management and the Transformation Model
Business EnglishUniversity StudentsWorking Adults
C12
What is Forecasting in Supply Chain Management
Business EnglishUniversity StudentsWorking Adults
✨ おすすめ動画
C11
75 文
How to EASILY Create a Driver-Based Forecast (FP&A)
Business EnglishUniversity StudentsWorking Adults
C12
236 文
The 7 Quality Control (QC) Tools Explained with an Example!
Business EnglishUniversity StudentsWorking Adults
C110
30 文
Research Source Mastery: How to Find & Verify Credible Data (2026)
Business EnglishUniversity StudentsWorking Adults
C11
62 文
English Job Interview Tips and Tricks 2 - How to Answer Job Interview Questions in English
Business EnglishInterview EnglishUniversity Students
C11
120 文
Learn Data Modeling in 8 minutes: Dimensional Data Modeling, Data Vault, and One Big Table
Business EnglishUniversity StudentsWorking Adults
C11
258 文
What is Accounting ? An Introduction to Accounting.
Business EnglishUniversity StudentsWorking Adults
C11
125 文
Introduction to statistical prediction models
Business EnglishUniversity StudentsWorking Adults
C18
45 文
Types of Surveys to Run Throughout the Design Process
Business EnglishUniversity StudentsWorking Adults
C13
444 文
Here's what's REALLY Happening...
Business EnglishUniversity StudentsWorking Adults
C128 文
GATHERING IN THE DIGITAL ERA
Business EnglishUniversity StudentsWorking Adults
C11
111 文
Financial Analysis Fundamentals: How to Analyze Financial Statements Like a Pro Part 7 | CFI Course
Business EnglishUniversity StudentsWorking Adults
この動画で鍛える英語スピーキングの目標
ビジネスシーンでよく使われる「予測手法」について話す際の、論理的な語り方と専門用語の使い方をマスターしましょう。定性的・定量的手法の違いを説明するだけでなく、具体的な例を挙げて分かりやすく伝えるスキルが身につきます。shadow speech（シャドーイング）で発音とリズムを磨き、自然な英語表現を即座に使えるようになりましょう！
フレーズバンク：すぐに使える表現
- 「There are two methods that are usually used when it comes to...」（...に関しては通常2つの方法が使われます）
- 「This is where we basically start at the bottom of...」（これは基本的に...の最下部から始めるところです）
- 「...is a great way to go when you have...」（...の場合、...するのが良い方法です）
- 「You could use... to do something like that」（そのようなことをするには...を使えます）
これらのフレーズを使えば、複雑な内容もスムーズに説明できます。英語シャドーイングで何度も繰り返し練習し、口癖のようになるまで慣らしましょう。
弱点を直そう：発音とリズムのポイント
この動画の中で頻出する「forecasting」「quantitative」「qualitative」などの長い単語は、発音が難しいかもしれません。特に「quantitative」の「tita」の部分は、舌を軽く動かして明確に発音しましょう。また、文の途中で息を吸うタイミングも重要です。shadow speakをする際は、話者の息継ぎや強調する単語を注意深く観察し、真似してみてください。これで英語の発音を良くするだけでなく、ネイティブ並みのリズムも身につきます。
少しずつでも毎日練習すれば、必ず進歩します。今日からこの動画を使って英語スピーキング練習を始めましょう！
この動画の文法
話し手がよく使っている文型を、動画の実際の表現とともに紹介します。
| 文型 | 動画での表現 |
|---|---|
| 受動態 be + 過去分詞 — 誰がするかより、何が起きるかに焦点を当てる | was done · was marketed · is called |
| 現在完了形 have/has + 過去分詞 — 過去の出来事が今も関係している | you've produced · I've broken · I've selected |
シャドーイングとは？英語上達に効果的な理由
シャドーイング（Shadowing）は、もともとプロの通訳者養成プログラムで開発された言語学習法で、多言語習得者として知られるDr. Alexander Arguelles によって広く普及されました。方法はシンプルですが非常に効果的：ネイティブスピーカーの英語を聞きながら、1〜2秒の遅延で声に出してすぐに繰り返す——まるで「影（shadow）」のように話者を追いかけます。文法ドリルや受動的なリスニングと異なり、シャドーイングは脳と口の筋肉が同時にリアルタイムで英語を処理・再現することを強制します。研究により、発音精度、抑揚、リズム、連音、リスニング力、そして会話の流暢さが大幅に向上することが確認されています。IELTSスピーキング対策や自然な英語コミュニケーションを目指す方に特におすすめです。