YouTube Transcript:
A11. Time Series

Skip watching entire videos - get the full transcript, search for keywords, and copy with one click.

Video Transcript

welcome back we will now discuss time

series as part of your data analysis

under analyze face time series is a

sequence of observations over regularly

spaced intervals of time as mentioned

the time series chart is used to track

the performance of your data or your

process over time example of

applications are monthly unemployment

rates for the previous five years daily

production at a manufacturing plant for

a month they get by decade population of

a state of the previous century it can

also include your SL age or service

level agreements which are monitored on

a monthly basis or the number of

customers that you have served for the

past weeks on a daily basis or on a

monthly basis or the number of incidents

of mishaps in an organization which is

recorded on a monthly basis as long as

you are trying to monitor or check the

performance of your data regardless of

its type either continuous or attribute

or discrete you can use time series as

long as your x-axis is a unit of time

now let's do on the anatomy of a time

series chart as you can see here we have

the y axis which corresponds to a

continuous variable which is gold price

on a dollar per ounce and your x axis

which is here as you can see here there

are data points that are plotted across

time and here are the time elements

remember that you can only use line

chart when you're using time series not

using bar child so that's the common

mistake that visual analysts do commit

using bar charts when their x axis is a

unit of time so always use the time

series chart on those particular

applications why do we use time series chart

chart

generally speaking we use time series

chart because we want to look for

patterns or we want to observe how our

process output behave across time we are

looking for what we call patterns which

can be further subdivided into 3 which

are trend

shift or cycle let's go with trend a

trend is a series of data points that is

increasing or decreasing increasing

the time decreasing attendance rate

increasing rejection rate so those are

possible scenarios that we can see next

is chef chef means an old performance

shifting to a new level of performance

either because of an improvement or

either because of a problem now as you

can see here there is a group of data

points here which shifted downward so

this is a particular example let's say

this is the current performance of your

attendance rate for the past weeks and

then it suddenly drop or this could be

the performance of your yield and it

suddenly drop due to something that

happened when you detect shift there has

been a sudden change of an element of

the process that triggered this

particular shift next cycle cycle as

represented by waves like this this

could also mean a CSUN

so when you say cycle it's a repetitive

cyclical movement of your data points

from one series of time periods to

another let's say holiday there could be

holiday season or cycle so those are

example of cycle though we can have as

we look on our time series chart now in

using time series chart this is the

important patterns that we have to look

for if in case we found out any of this

we should ask ourselves what happened to

our process why is it giving us this

kind of patterns let's take our first

example as we use time series flood the

manager of a shipping yard wants to

study the amount of cargo that is

transported the manager collects the

weight of all cargo that passes through

the shipping yard each month now we will

go to our source sheet and copy our data

shipping to our Minitab so let's look

for our shipping data and our source

worksheet just have to copy this and

then go to our Minitab again you have to

paste your data on the first row here

without the CRO numbers now we have two

columns the month running through more

than 50 and 100 seven months of data and

this is the weight of the material ship

when using time series we can go either stat

stat

we have an option here time-series or we

can go to graph we have an option here

so for this case we'll go to graph again

time series plot plot the data in the

order that it appears in the worksheet

okay so whatever you are giving the

Minotaur software it will just plot it

based on the order so let's go to graph

and then click time series let's use

simple because we have one set of data

here one column and then you just have

to click your weight data here which

data columns do you want operate a time

series plot so double left-click that

particular column for this case C 2

double left-click and then you just have

to click OK

okay so we have now a time series of

weight across 107 months so this is your

y axis corresponding to your rate data

across 107 months and these are the data

plotted all over those periods do you

see any patterns okay so for this case

we have this lump of data points here

I think this would mean a shift from

here to here okay and there's a natural

cycle as you can see several number of

trends decreasing trend we have

increasing trend increasing trend

decreasing trend those patterns are

available now what we want to do is to

understand the behavior of our process

why it exhibits this patterns and make

it as a basis for us to further

investigate now when you're using time

series for your hypothesis testing or

root-cause validation you just have to

state your hypothesis and if your data

and if you want to prove that using your

data and check it across time then you

can use time series let's say is there

any dependency of your problem - let's

say the day of the week or week of the

month now you can use time series to

prove that so that's the first example

let's go to the next example the

administrator of a hospital wants to

examine the number of courage' patients

admitted over the past 24 months to

analyze trends in the

so for this case we will be using the

patient data so let's go to our

worksheet patient and then let's copy

the data let's go back Minitab create

new worksheet by pressing ctrl n and

close this graph now we can paste a data

so we have two columns again this time

it's indicated here on the first call of

the month but this data we only get the

first 12 months of a year we didn't

include the second set of twelve months

and this column corresponds the number

of cards of patients now what we want to

do is to create a time series chart

graph time series simple let's erase

this and replace with number of patients

now before you click OK you can actually

specify the time for scale because you

have C one month for you to replace the

index one to twelve we can click

timescale and then click stamp click on

this dialog box here and then the

available stamp column will appear so

you want to stomp the x-axis with the

month corresponding to the data points

and then you can click OK and then click

ok now we have your time series chart of

the number of patient cards of patients

over one year now do you see patterns

yes there is an increasing pattern here

from October November and December and

there is a decreasing pattern here from

January February March and April there

could be a possible shift this lump of

data going to this lump of data and

another shift here there could be cycle

if we replicate this one we move to the

next year we can see cycle now the

question is why is there this particular

behavior of this process or output data

this is number of cards of patients over

a year you can see that in November and

December there has been an increase

until January February and March

it's because do you have any guess yes

it's because of the festivities or the

holiday so this is a data from the US we

have November December those months

until March those are festivities we

just the reason why the number of cards

of patients should up on this month by

understanding how the process output

behave over time can give us insights

and how we can do our actions or

improvement actions moving forward

validating also our claims if let's say

is months a significant factor on the

number of courage of patients if that's

your hypothesis or is the number of

project patients dependent on the month

if that's the case then the answer will

be yes as seen in this particular output

chart that we have moving forward this

is how you will validate your hypothesis

if you have a continuous data or an

attribute data as your Y and a unit of

Click on any text or timestamp to jump to that moment in the video

Most transcripts ready in under 5 seconds

One-Click Copy125+ LanguagesSearch ContentJump to Timestamps

Paste YouTube URL

Enter any YouTube video link to get the full transcript

Most transcripts ready in under 5 seconds

Get Our Chrome Extension

Get transcripts instantly without leaving YouTube. Install our Chrome extension for one-click access to any video's transcript directly on the watch page.

Add to Chrome — Free

Works with YouTube, Coursera, Udemy and more educational platforms

Get Instant Transcripts: Just Edit the Domain in Your Address Bar!

YouTube

←

→

↻

https://www.youtube.com/watch?v=UF8uR6Z6KLc

YoutubeToText

←

→

↻

https://youtubetotext.net/watch?v=UF8uR6Z6KLc

YouTube Transcript:A11. Time Series

Video Transcript

Paste YouTube URL

Transcript Extraction Form

Get Our Chrome Extension

Get Instant Transcripts: Just Edit the Domain in Your Address Bar!

YouTube Transcript:
A11. Time Series