| Course title | |||||
| 情報処理学 [Computer Science] | |||||
| Course category | Requirement | Credit | 2 | ||
| Department | Year | 1～ | Semester | 1st | |
| Course type | 1st | Course code | 01rn1011a | ||
| Instructor(s) | |||||
| 酒井 憲司, 小川 智浩, 島谷 健一郎 [SAKAI Kenshi, OGAWA Tomohiro, SHIMATANI Kenichiro] | |||||
| Facility affiliation | Faculty of Agriculture | Office | afjgxte/L1151 | Email address | |
| Course description |
|
Learn the basic concepts of statistical models that make estimations and predictions from limited data, and the necessary mathematical concepts.10 lectures and some exercises will be in the meantime. Through the regression model that analyzes the relationship between two variables and the generalized linear model that is the development version, learn the estimation method of unknown parameters, the basis of prediction and estimation by statistical model, and the statistical hypothesis test. We will also demonstrate demos such as random number generation and checking and prediction of data analysis results by simulation. These will be given in the first 10 lectures. Students will learn about data processing methods such as statistical processing and signal processing methods required to extract significant information from data obtained by field surveys / experiments etc. in the Department of Regional Ecosystems, through exercises. We also learn basics of programming with exercises through learning of useful simulations and tools for problem solving / hypothesis preliminary verification. The last 5 lectures is taught by a teacher who has practical experience (Ogawa). The instructor in charge is engaged in design / development work using information processing technology such as simulation in a company, and learns the basics of data processing / simulation programming often used in the process of design / development in class. |
| Expected Learning |
|
Achievement standard (the first half) 1. To be able to answer the question why statistical analysis is first needed for data obtained by experiments and surveys. 2. It will be possible to distinguish between findings that can not be obtained without statistical analysis and estimations and forecasts that can not be achieved with data alone even if statistical analysis is performed. 3. From numerical results of statistical analysis, it becomes possible to interpret while knowing what and how far. Achievement standard (the second half) 1 Be able to perform data processing such as statistical calculation and signal processing using spreadsheet software. 2 Be able to understand the concept of simulation of dynamic model. 3 Be able to program after understanding the algorithm. |
| Course schedule |
|
first half Proceed in the following order. The corresponding textbook page will be presented during the first class. 1. Examples of statistical analysis of biological data, how to view and visualize data, tables and graphs, aggregate statistics, short history of statistics 2. Generalized linear models (linear regression, logistic regression) that capture changes and relationships and their basic concepts: probability distribution, expectation value, variance, statistical model, likelihood, maximum likelihood method, least squares method 3. Random number generation, prediction by simulation, confirmation of data reproducibility and model diagnosis 4. Analysis of variance, significance test and hypothesis test, contingency table and chi-square test 5. Statistical Thinking: Collective Thinking and Inductive Inference Latter half 1. With exercises about data processing such as statistical processing and signal processing using spreadsheet software Excel . 2. Learn about dynamic models and the concept of simulation with exercises. 3.Programming exercise |
| Prerequisites |
| This lecture will proceed with the basic operation of the personal computer and the handling of files as if you have already learned in "Information processing class" in high school. |
| Required Text(s) and Materials |
|
自然科学の統計学 The University of Tokyo Press The content of learning is presented on Moodle, and changes may be made as the course progresses. In addition, please download problems, notes, etc. individually by downloading from Moodle. |
| References |
|
「統計学がよくわかる本―Excel解説付き」宮川庸一著 アイ・ケイコーポレーション、「統計思考の世界」三中信宏著 技術評論社、 「フィールドデータによる統計モデリングとAIC」島谷健一郎著、近代科学社. If necessary, we will introduce general books on Excel, programming, simulation, etc. in the course. |
| Assessment/Grading |
| Small exam (20%) and final exam (20%), homework given during class (20%), report (40%). |
| Message from instructor(s) |
|
If you can not follwo mathematics in the textbook, learn in advance that "Statistics can be understood well-with Excel commentary" mentioned in the reference, and have some background knowledge before going to class. Internet, television, newspapers, etc., everyday words such as artificial intelligence (AI), big data, data scientists, machine learning, etc. have come out. It is recommended that you be aware of the relationship between these articles and the “statistics” you will learn in class. The second half will be exercise-based classes. We will mainly evaluate the grades based on the submitted items. |
| Course keywords |
| Statistical model, probability distribution, regression model, generalized linear model, random number generation, simulation Excel, programming, simulation |
| Office hours |
| Kan?nakagiri zuiji m?ru nite tai? shimasu. 19/5000 We will respond by email as often as possible. |
| Remarks 1 |
| Remarks 2 |
| Related URL |
| Lecture Language |
| Japanese |
| Language Subject |
| Last update |
| 3/4/2021 1:54:40 PM |