For example, data analysis thinking, structured thinking, formulaic thinking and learning method system thinking can help you analyze from a certain angle and keep clear logic even if you encounter unfamiliar problems;
A certain business understanding ability, able to understand the business ideas behind the business. Only by understanding the problem can it be transformed into a problem of data analysis, and can we know how to set the analysis goal and analyze it.
Basic theoretical knowledge: mathematical statistics, model principles, recent market research, etc.
Use of routine analysis tools: common office software (Excel, PPT, mind map), database, statistical analysis tools, data mining, etc.
Ability to report and visualize data. No matter how good the data analysis is, if it can't be "expressed" in a concise and easy-to-understand way, the result will be greatly reduced.
Wait, wait, wait. ......
So how should we improve these abilities? Let's talk specifically about how to lay these basic strengths.
Starting with analytical theory and tool practice
1, analytical theory
Analysis theory includes: defining business scenarios, determining analysis objectives, constructing analysis system, and sorting out core indicators.
What we need to do is, first, to clarify what kind of business scenarios, and different businesses have different analysis systems; Then, combined with business problems, determine the analysis objectives, list the core indicators, and then collect and sort out the required data.
Recommended books: data management, campaign big data.
Several steps of data analysis:
(1) data acquisition
The acquisition of data often seems simple, but analysts need to understand the problems in the business, that is, to turn the problems into data problems to solve, such as what data are needed, from which angles to analyze, and then collect data after clarifying these problems.
This link requires data analysts to have structured logical thinking.
Recommended books: The Golden Pagoda Principle and McKinsey Trilogy: McKinsey Consciousness, Tools and Methods.
Recommended tools: mind mapping tools (Xmind Baidu brain map, etc. )
(2) Data processing
Data processing needs to master efficient tools:
Excel and high-end skills:
Basic operation, function formula, pivot table, VBA program development.
I usually go through the basics first, know what is what, and then find some cases to practice. Visit the excelhome forum more often, usually think about how to solve problems with excel, make good use of plug-ins and remember to save them.
Professional reporting tools:
(Large enterprises will use it) You can design a general template for daily report making, and you can get started as long as you can write SQL.
Compared with excel, this tool has lower technical requirements, and can quickly develop regular reports and dynamic reports.
Use of the database:
Proficient in SQL language (very important! ! ! ), the common ones are Oracle, SQL sever, My SQL, etc.
Learning hadoop and other popular distributed databases and improving personal ability will be helpful for job hunting.
(3) data analysis
Analysis of data often requires various statistical analysis models, such as association rules, clustering, classification, prediction models and so on.
Therefore, it is inevitable to master some statistical analysis tools:
LPSS series: the old statistical analysis software, SPSS Statistics (partial statistical function, market research) and SPSS Modeler (partial data mining), does not need programming and is easy to learn.
SAS: Classic mining software, which needs programming.
R: Open source software, a new and popular software, is more efficient in dealing with unstructured data and needs programming.
Various BI tools: Tableau, PowerBI, FineBI, can freely visualize and analyze the processed data, and the chart effect is amazing.
Recommended books:
"Say a novice can't analyze data" series, introductory book, most suitable for novices.
The actual combat, ideas, methods, skills and applications of data mining and data operation are very systematic and comprehensive.
Quantitative analysis method and application of market research, concise editor, Renmin University of China Press.
(4) data visualization
Many data analysis tools have covered data visualization, and only need to effectively present and report data results, which can be displayed in word\PPT\H5.
2. Tool practice
(1) getting started with Xiaobai, it is recommended to start with Excel tools. Take Excel as an example here:
Learning Excel is a gradual process;
Basic: Simple table data processing, printing, querying, filtering and sorting.
Functions and formulas: commonly used functions, advanced data calculation, array formulas, multidimensional references and functions.
Visual charts: graphic icon display, advanced charts and chart plug-ins.
Development of Pivottable and VBA Program ......
Visit the excelhome forum more, think more about how to solve problems with excel, and learn to use various plug-ins, which will help you to use Excel skillfully.
Among them, function and PivotTable are two key points.
function
Excel functions that must be mastered when making data templates;
Date function: Date, month, year, date, today, working day and week date functions are necessary for making analysis templates. You can use the date function to control the display of data and query the data within a specified time period.
Mathematical functions: product, rand, randbetween, round, sum, sumif, sumifs, sumproduct.
Statistical functions: large, small, max, min, median, mode, rank, count, countif, countifs, average, averageif, averageifs statistical functions play an important role in data analysis. Average, maximum, median and modulus are all used.
Find and reference functions: choose, match, index, indirect, column, row, vlookup, hlookup, lookup, offset, getpivotdata. Needless to say, vlookup, in particular, will hardly complicate this function.
Text functions: find, search, text, value, concatenate, left, right, mid and len are mainly used in the data sorting stage.
Logical functions: and, or, false, true, if, iferror.
(The above society can basically kill 90% of office white-collar workers! )
PivotTable
The function of pivot table is to generate interactive reports from a large number of data, which has some important functions: classified summary, average, maximum and minimum values, automatic sorting, automatic screening and automatic grouping; Can analyze proportion, year-on-year, quarter-on-quarter, fixed ratio, user-defined formula and so on.
In reality, taking data or reports +EXCEL+PPT seems to be the mainstream form.
Tools, whether business personnel or analysts, can use automatic data retrieval tools or BI tools to make reports, reducing the time of repeated operations.
Secondly, increase communication with business personnel to fully understand business needs. When your business level is similar to or even higher than theirs, you will naturally know what the real needs behind their words are.
Finally, from a higher perspective, the basic granularity of statements is indicators, which can sort out the basic indicator system of enterprises, make statements from the perspective of business analysis, standardize the work of statements, reduce the redundancy of statements and avoid making a statement easily. Standardization includes index classification, index naming, business caliber, technical caliber, implementation method and so on. In fact, the ultimate goal is to achieve the consistency of report data, reduce repeated report development and reduce system overhead.
In my spare time, I can supplement my knowledge of mathematical statistics, learn R and Python languages, learn common mining models, and develop into a senior analyst!
Come on, duck!
The above is today's sharing. The ability of data analysis sounds great and abstract. Although it is soft power, it is a hard requirement of the industry! Quantitative change leads to qualitative change, step by step, in order to achieve analogy, the project will become more and more convenient.