Data Analysis (Head First) (1 Original)

個数:

Data Analysis (Head First) (1 Original)

  • 提携先の海外書籍取次会社に在庫がございます。通常2週間で発送いたします。
    重要ご説明事項
    1. 納期遅延や、ご入手不能となる場合が若干ございます。
    2. 複数冊ご注文の場合、分割発送となる場合がございます。
    3. 美品のご指定は承りかねます。
  • ≪洋書のご注文につきまして≫ 「海外取次在庫あり」および「国内仕入れ先からお取り寄せいたします」表示の商品でも、納期の目安期間内にお届けできないことがございます。あらかじめご了承ください。

  • 製本 Paperback:紙装版/ペーパーバック版/ページ数 445 p.
  • 言語 ENG,ENG
  • 商品コード 9780596153939
  • DDC分類 005.74

Full Description


Today, interpreting data is a critical decision-making factor for businesses and organizations. If your job requires you to manage and analyze all kinds of data, turn to "Head First Data Analysis", where you'll quickly learn how to collect and organize data, sort the distractions from the truth, find meaningful patterns, draw conclusions, predict the future, and present your findings to others. Whether you're a product developer researching the market viability of a new product or service, a marketing manager gauging or predicting the effectiveness of a campaign, a salesperson who needs data to support product presentations, or a lone entrepreneur responsible for all of these data-intensive functions and more, the unique approach in "Head First Data Analysis" is by far the most efficient way to learn what you need to know to convert raw data into a vital business tool.You'll learn how to: determine which data sources to use for collecting information; assess data quality and distinguish signal from noise; build basic data models to illuminate patterns, and assimilate new information into the models; cope with ambiguous information; design experiments to test hypotheses and draw conclusions; use segmentation to organize your data within discrete market groups; visualize data distributions to reveal new relationships and persuade others; predict the future with sampling and probability models; clean your data to make it useful; and, communicate the results of your analysis to your audience. Using the latest research in cognitive science and learning theory to craft a multi-sensory learning experience, "Head First Data Analysis" uses a visually rich format designed for the way your brain works, not a text-heavy approach that puts you to sleep.

Table of Contents

introduction to data analysis
Break it down
Data is everywhere
Acme Cosmetics needs your help 2 (1)
The CEO wants data analysis to help 3 (1)
increase sales
Data analysis is careful thinking about 4 (1)
evidence
Define the problem 5 (1)
Your client will help you define your 6 (2)
problem
Acme's CEO has some feedback for you 8 (1)
Break the problem and data into smaller 9 (1)
pieces
Now take another look at what you know 10 (3)
Evaluate the pieces 13 (1)
Analysis begins when you insert yourself 14 (1)
Make a recommendation 15 (1)
Your report is ready 16 (1)
The CEO likes your work 17 (1)
An article just came across the wire 18 (2)
You let the CEO's beliefs take you down 20 (1)
the wrong path
Your assumptions and beliefs about the 21 (1)
world are your mental model
Your statistical model depends on your 22 (3)
mental model
Mental models should always include 25 (1)
what you don't know
The CEO tells you what he doesn't know 26 (2)
Acme just sent you a huge list of raw 28 (3)
data
Time to drill further into the data 31 (1)
General American Wholesalers confirms 32 (3)
your impression
Here's what you did 35 (1)
Your analysis led your client to a 36 (2)
brilliant decision
experiments
Test your theories
Can you show what you believe?
It's a coffee recession! 38 (1)
The Starbuzz board meeting is in three 39 (2)
months
The Starbuzz Survey 41 (1)
Always use the method of comparison 42 (1)
Comparisons are key for observational 43 (1)
data
Could value perception be causing the 44 (2)
revenue decline?
A typical customer's thinking 46 (1)
Observational studies are full of 47 (1)
confounders
How location might be confounding your 48 (2)
results
Manage confounders by breaking the data 50 (3)
into chunks
It's worse than we thought! 53 (1)
You need an experiment to say which 54 (1)
strategy will work best
The Starbuzz CEO is in a big hurry 55 (1)
Starbuzz drops its prices 56 (1)
One month later... 57 (1)
Control groups give you a baseline 58 (3)
Not getting fired 101 61 (1)
Let's experiment for real! 62 (1)
One month later... 63 (1)
Confounders also plague experiments 64 (1)
Avoid confounders by selecting groups 65 (2)
carefully
Randomization selects similar groups 67 (1)
Randomness Exposed 68 (3)
Your experiment is ready to go 71 (1)
The results are in 72 (1)
Starbuzz has an empirically tested 73 (3)
sales strategy
optimization
Take it to the max
We all want more of something
You're now in the bath toy game 76 (3)
Constraints limit the variables you 79 (1)
control
Decision variables are things you can 79 (1)
control
You have an optimization problem 80 (1)
Find your objective with the objective 81 (1)
function
Your objective function 82 (1)
Show product mixes with your other 83 (1)
constraints
Plot multiple constraints on the same 84 (1)
chart
Your good options are all in the 85 (2)
feasible region
Your new constraint changed the 87 (3)
feasible region
Your spreadsheet does optimization 90 (4)
Solver crunched your optimization 94 (3)
problem in a snap
Profits fell through the floor 97 (1)
Your model only describes what you put 98 (1)
into it
Calibrate your assumptions to your 99 (4)
analytical objectives
Watch out for negatively linked 103(5)
variables
Your new plan is working like a charm 108(1)
Your assumptions are based on an 109(3)
ever-changing reality
data visualization
Pictures make you smarter
You need more than a table of numbers
New Army needs to optimize their website 112(1)
The results are in, but the information 113(1)
designer is out
The last information designer submitted 114(1)
these three infographics
What data is behind the visualizations? 115(1)
Show the data! 116(1)
Here's some unsolicited advice from the 117(1)
last designer
Too much data is never your problem 118(1)
Making the data pretty isn't your 119(1)
problem either
Data visualization is all about making 120(3)
the right comparisons
Your visualization is already more 123(1)
useful than the rejected ones
Use scatterplots to explore causes 124(1)
The best visualizations are highly 125(1)
multivariate
Show more variables by looking at 126(4)
charts together
The visualization is great, but the web 130(1)
guru's not satisfied yet
Good visual designs help you think 131(1)
about causes
The experiment designers weigh in 132(3)
The experiment designers have some 135(1)
hypotheses of their own
The client is pleased with your work 136(1)
Orders are coming in from everywhere! 137(3)
hypothesis testing
Say it ain't so
The world can be tricky to explain
Gimme some skin... 140(1)
When do we start making new phone skins? 141(1)
PodPhone doesn't want you to predict 142(1)
their next move
Here's everything we know 143(1)
ElectroSkinny's analysis does fit the 144(1)
data
ElectroSkinny obtained this 145(1)
confidential strategy memo
Variables can be negatively or 146(3)
positively linked
Causes in the real world are networked, 149(1)
not linear
Hypothesize PodPhone's options 150(1)
You have what you need to run a 151(1)
hypothesis test
Falsification is the heart of 152(8)
hypothesis testing
Diagnosticity helps you find the 160(3)
hypothesis with the least
disconfirmation
You can't rule out all the hypotheses, 163(1)
but you can say which is strongest
You just got a picture message... 164(3)
It's a launch! 167(3)
bayesian statistics
Get past first base
You'll always be collecting new data
The doctor has disturbing news 170(3)
Let's take the accuracy analysis one 173(1)
claim at a time
How common is lizard flu really? 174(1)
You've been counting false positives 175(1)
All these terms describe conditional 176(1)
probabilities
You need to count false positives, true 177(1)
positives, false negatives, and true
negatives
1 percent of people have lizard flu 178(3)
Your chances of having lizard flu are 181(1)
still pretty low
Do complex probabilistic thinking with 182(1)
simple whole numbers
Bayes' rule manages your base rates 182(1)
when you get new data
You can use Bayes' rule over and over 183(1)
Your second test result is negative 184(1)
The new test has different accuracy 185(1)
statistics
New information can change your base 186(3)
rate
What a relief! 189(3)
subjective probabilities
Numerical belief
Sometimes, it's a good idea to make up
numbers
Backwater Investments needs your help 192(1)
Their analysts are at each other's 193(5)
throats
Subjective probabilities describe 198(1)
expert beliefs
Subjective probabilities might show no 199(2)
real disagreement after all
The analysts responded with their 201(1)
subjective probabilities
The CEO doesn't see what you're up to 202(5)
The CEO loves your work 207(1)
The standard deviation measures how far 208(5)
points are from the average
You were totally blindsided by this news 213(4)
Bayes' rule is great for revising 217(6)
subjective probabilities
The CEO knows exactly what to do with 223(1)
this new information
Russian stock owners rejoice! 224(2)
heuristics
Analyze like a human
The real world has more variables than
you can handle
LitterGitters submitted their report to 226(1)
the city council
The LitterGitters have really cleaned 227(1)
up this town
The LitterGitters have been measuring 228(1)
their campaign's effectiveness
The mandate is to reduce the tonnage of 229(1)
litter
Tonnage is unfeasible to measure 230(1)
Give people a hard question, and 231(1)
they'll answer an easier one instead
Littering in Dataville is a complex 232(1)
System
You can't build and implement a unified 233(3)
litter-measuring model
Heuristics are a middle ground between 236(3)
going with your gut and optimization
Use a fast and frugal tree 239(1)
Is there a simpler way to assess 240(4)
LitterGitters' success?
Stereotypes are heuristics 244(2)
Your analysis is ready to present 246(3)
Looks like your analysis impressed the 249(3)
city council members
histograms
The shape of numbers
How much can a bar graph tell you?
Your annual review is coming up 252(2)
Going for more cash could play out in a 254(1)
bunch of different ways
Here's some data on raises 255(7)
Histograms show frequencies of groups 262(1)
of numbers
Gaps between bars in a histogram mean 263(1)
gaps among the data points
Install and run R 264(1)
Load data into R 265(1)
R creates beautiful histograms 266(5)
Make histograms from subsets of your 271(5)
data
Negotiation pays 276(1)
What will negotiation mean for you? 277(3)
regression
Prediction
Predict it
What are you going to do with all this 280(3)
money?
An analysis that tells people what to 283(1)
ask for could be huge
Behold... the Raise Reckoner! 284(2)
Inside the algorithm will be a method 286(6)
to predict raises
Scatterplots compare two variables 292(2)
A line could tell your clients where to 294(3)
aim
Predict values in each strip with the 297(1)
graph of averages
The regression line predicts what 298(2)
raises people will receive
The line is useful if your data shows a 300(4)
linear correlation
You need an equation to make your 304(2)
predictions precise
Tell R to create a regression object 306(3)
The regression equation goes hand in 309(1)
hand with your scatterplot
The regression equation is the Raise 310(3)
Reckoner algorithm
Your raise predictor didn't work out as 313(3)
planned...
error
Err well
The world is messy
Your clients are pretty ticked off 316(1)
What did your raise prediction 317(1)
algorithm do?
The segments of customers 318(3)
The guy who asked for 25% went outside 321(1)
the model
How to handle the client who wants a 322(5)
prediction outside the data range
The guy who got fired because of 327(1)
extrapolation has cooled off
You've only solved part of the problem 328(1)
What does the data for the screwy 329(1)
outcomes look like?
Chance errors are deviations from what 330(4)
your model predicts
Error is good for you and your client 334(1)
Chance Error Exposed 335(1)
Specify error quantitatively 336(1)
Quantify your residual distribution 337(1)
with Root Mean Squared error
Your model in R already knows the 338(2)
R.M.S. error
R's summary of your linear model shows 340(6)
your R.M.S. error
Segmentation is all about managing error 346(4)
Good regressions balance explanation 350(2)
and prediction
Your segmented models manage error 352(5)
better than the original model
Your clients are returning in droves 357(3)
relational databases
Can you relate?
How do you structure really, really
multivariate data?
The Dataville Dispatch wants to analyze 360(1)
sales
Here's the data they keep to track 361(1)
their operations
You need to know how the data tables 362(3)
relate to each other
A database is a collection of data with 365(1)
well-specified relations to each other
Trace a path through the relations to 366(1)
make the comparison you need
Create a spreadsheet that goes across 366(5)
that path
Your summary ties article count and 371(3)
sales together
Looks like your scatterplot is going 374(1)
over really well
Copying and pasting all that data was a 375(1)
pain
Relational databases manage relations 376(1)
for you
Dataville Dispatch built an RDBMS with 377(2)
your relationship diagram
Dataville Dispatch extracted your data 379(3)
using the SQL language
Comparison possibilities are endless if 382(1)
your data is in a RDBMS
You're on the cover 383(3)
cleaning data
Impose order
Your data is useless...
Just got a client list from a defunct 386(1)
competitor
The dirty secret of data analysis 387(1)
Head First Head Hunters wants the list 388(4)
for their sales team
Cleaning messy data is all about 392(1)
preparation
Once you're organized, you can fix the 393(1)
data itself
Use the # sign as a delimiter 394(1)
Excel split your data into columns 395(4)
using the delimiter
Use Substitute to replace the carat 399(1)
character
You cleaned up all the first names 400(2)
The last name pattern is too complex 402(1)
for Substitute
Handle complex patterns with nested 403(1)
text formulas
R can use regular expressions to crunch 404(2)
complex data patterns
The sub command fixed your last names 406(1)
Now you can ship the data to your client 407(1)
Maybe you're not quite done yet... 408(1)
Sort your data to show duplicate values 409(3)
together
The data is probably from a relational 412(1)
database
Remove duplicate names 413(1)
You created nice, clean, unique records 414(1)
Head First Head Hunters is recruiting 415(1)
like gangbusters!
Leaving town... 416(2)
leftovers
The Top Ten Things (we didn't cover)
You've come a long way
Everything else in statistics 418(1)
Excel skills 419(1)
Edward Tufte and his principles of 420(1)
visualization
PivotTables 421(1)
The R community 422(1)
Nonlinear and multiple regression 423(1)
Null-alternative hypothesis testing 424(1)
Randomness 424(1)
Google Docs 425(1)
Your expertise 426(2)
install r
Start R up!
Behind all that data-crunching power is
enormous complexity
Get started with R 428(4)
install excel analysis tools
The ToolPak
Some of the best features of Excel aren't
installed by default
Install the data analysis tools in Excel 432