馃攳
The Fundamentals of Predictive Analytics - Data Science Wednesday - YouTube
Channel: unknown
[8]
hello and welcome back to data science
Wednesday my name is Tessa Jones and I'm
[13]
a data scientist with decisive data and
today we're going to talk about
[17]
predictive analytics and what it can do
for you
[20]
predictive analytics fits into the
spectrum of analytics that we've talked
[24]
about before starting with descriptive
which is the most basic of the analytics
[29]
it is basically just cleaning relating
summarizing and visualizing your data
[34]
really getting to the questions about
what's happening in my business and then
[39]
there's diagnostic which is really
getting down to why things are happening
[43]
what's causing my revenue to decline or
to increase how are things related
[48]
things like that so if you've got a good
base in both of these then we're ready
[52]
to move into predictive analytics which
is gonna dive into what's gonna happen
[57]
in the future which is super powerful if
you're a business person and you want to
[62]
be able to make good business questions
if you have at least an idea of what
[66]
might happen in the future your your
answers are already gonna be a little
[70]
bit better so let's dive in so let's go
with an example cuz that just makes it
[76]
easier to kind of flow through what's
actually happening here so let's pretend
[80]
that we are grocery store owners and if
we're already talking about predictive
[85]
analytics you should have a pretty good
grasp on descriptive and predictive and
[90]
diagnostic analytics so you probably
already have a decent dashboard that
[94]
really tells you what's happening in
your business right now so something
[98]
like this where you have you know
something here that tells you revenue by
[102]
different departments like foods meats
or foods and pastry or how your sales
[109]
changes by product or overtime things
like that so you have an idea of what's
[114]
happening in your business but now you
really want to know what's going to
[117]
happen in my business so one really
common question is how many of a given
[121]
product am I going to sell for every
store because this can really give you
[126]
quite answer questions around how you're
going to support supply chain processes
[130]
or how you're going to manage the
profits that you're going to have things
[135]
like that so the first thing we need to
do is talk about what happened
[139]
past we really can't do anything or
predict very easily unless we know or at
[144]
least have an idea of what's happened in
the past so here we have three lines in
[149]
black that represent basically
historical data each line here is one
[154]
year worth of sales for a given product
and then the green line here is the
[159]
current year and here's today and if we
build a predictive model it's going to
[163]
tell us what's going to happen for the
rest of the year so if this is all set
[168]
up and we build a model basically we mix
this information with all the data
[174]
that's really clean and well-organized
we mash it together with a bunch of
[179]
mathematics and coding and basically we
pop out some results and it shows up in
[185]
a visual like this where you have these
are the cells that we have had and these
[189]
are the cells that we think we're going
to have so a business person can look at
[193]
this chart and say wow we need to put a
lot more products to this store because
[198]
I see sales are going to increase or our
profit margins are going to be way
[202]
higher than we thought so we can start a
new program things like that you can
[207]
really start to get innovative with your
business decisions so let's pretend
[211]
we've built this model and it's been
running for a year and now we want to
[216]
know how well is this model actually
performing so down here we have a chart
[221]
that shows in black what we actually
sold and then in green what we thought
[225]
we were going to sell and we see that
there's some a couple of pretty big
[229]
misses right here we sold way more than
we thought we would which leaves risk to
[234]
you know missing out on inventory or
here we predicted we would sell more way
[239]
more than we did so both of these are
kind of misses and so we need to go back
[244]
and look at the data and understand what
assumptions we we applied that we're
[249]
maybe a little bit wrong or applied
incorrectly or look at the data maybe we
[254]
weren't accounting for something and we
kind of reorganize that and incorporate
[260]
it into the model and then we redeploy
it and then the then we have a better
[264]
model this cycle can you know happen a
couple times or it can happen many
[270]
it really depends on the data it really
depends on the objective it depends on a
[276]
lot of different things but we do try to
minimize the number of times that we're
[279]
having to iterate through this before we
can have a really sound predictive model
[283]
so that's predictive analytics in a
nutshell basically once you have a solid
[291]
foundation of descriptive and diagnostic
analytics we can really start pushing
[295]
forward with predictive analytics and
then next week we're going to start
[299]
talking about prescriptive analytics
which really gets to the questions of
[303]
okay now that we know what's happening
in the future what do we do about it I'm
[308]
Tessa Jones and that's a reindeer
Most Recent Videos:
You can go back to the homepage right here: Homepage