Spatial Economics & Machine Learning Theory: Intensive Seminar (2018)
From June to July this year, we held an intensive seminar to learn basic theories of spatial economics and machine learning. Here are the materials used in the seminar.
Spatial Economics Seminar
Based on: “The Spatial Economy: Cities, Regions, and International Trade” (Masahisa Fujita, Paul Krugman, Anthony J. Venables [author], Hiroyuki Koide [translator])
Chp. 1-3: Introduction, Urban Economics and Regional Science ●Chp.1-3.pdf
Chp. 4-5: The Dixit-Stiglitz Model of Monopolistic Competition and Its Spatial Implications, Core and Periphery ●Chp.4-5.pdf
Chp. 6-8: Many Regions and Continuous Space, Agricultural Transport Costs, Spatial Models of Urban Systems ●Chp.6-8.pdf
Chp. 9-10: The Monocentric Economy, The Emergence of New Cities ●Chp.9-10.pdf
Chp. 11-13: Evolution of a Hierarchical Urban System, An Empirical Digression, Ports, Transportation Hubs, and City Location ●Chp.11-13.pdf
Machine Learning Seminar
Deep Learning
Based on: “An Introduction to Deep Learning” (Masato Taki), “Deep Learning” (Takayuki Okatani) Deep Learning.pdf
Submodular function
Based on: “Machine learning with submodular functions” (Yoshinobu Kawahara, Kiyohito Nagano)
Co-clustering
Based on:”Relational Data Learning” (Katsuhiko Ishiguro, Kohei Hayashi)
Basic and Programming Seminars on Theory of Transportation (Y2018)
As a start-up, Fukuda Laboratory conducts a basic seminar on the three major basic theories required for traffic analysis in the first 2-3 months of the academic year. In 2018, the seminar aimed to learn the basic theories and application methods (programming skills) for two topics of “travel behavior analysis” and “traffic network analysis”.
Textbooks used:
- “Analysis and Modeling of Traffic Behavior: Theory/Model/Survey/Application” (Ryuichi Kitamura and Takayuki Morikawa [editors])
- “Equilibrium Analysis of Transportation Networks” (JSCE [editor])
Programming languages used:
- Python (some functions in Pythonbiogeme)
- R
【Intoduction】
Chp.1 Data Aggregation and Visualization 01Py.pdf 01R.pdf 01code.zip
Chp.2 Big Data Processing 02Py&R.pdf 02code_R&Py.zip
Spatial data analysis using QGIS 02QGIS.pdf 02QGIS.zip
Chp.3 Regression Analysis and Causal Effect Analysis 03R_Regression Analysis.pdf 03_Causal Effect Analysispdf 03code.zip
Chp.4 Creating documents and slides using LatexLatex, Bibliography Management with BibTeX and Zotero 04LaTex.pdf 04BibTex.pdf 04code.zip
【Travel Behavior Model】
Chp.5 Derivation and Parameter Estimatio of Multinomial Logit (MNL) Model 05Theory.pdf 05R.pdf 05Code_R.zip
Chp.6 Nested Logit Model (NL), Mixed Logit Model (MXL) and Parameter Estimation 06R.pdf 06Pybiogeme.pdf 06code.zip
Chp.7 Metropolitan Area Rail Demand Forecasting 07Railways.pdf Estimation of value of time 07Pybiogeme.pdf 07code.zip
Chp.8 Multi-Agent Simulation 08MATsim.pdf 08RL_R&Py.pdf 08code.zip
【Traffic Network Analysis】
Chp.9 Network Representation and Shortest Path Search 09 Network Representation.pdf 09Shortest path.pdf 09code.zip
Chp.10 Deterministic User Equilibrium Model and Calculation Algorithm 10R.pdf 10Py.pdf 10code.zip
Chp.11 Stochastic User Equilibrium Models and Markov Theory 11Stochastic.pdf Variant Demand UE 11Variant-Demand UE.pdf
Chp.12 Congestion Charging and Network Model 12 First Half.pdf 12 Second Half.pdf
※The pdf files marked “Py”, “Pybiogeme”, “R”, and “MATsim” are materials that include explanations of the program code, while the other materials are explanations of the theory and how to use the analysis software.
Basic/Programming Seminar on Transportation Theory (Y2017)
As a start-up, Fukuda Laboratory conducts a basic seminar on the three major basic theories required for traffic analysis in the first 2-3 months of the academic year. In 2017, the seminar aimed at to learn the basic theories and application methods (programming skills) for the two topics of “travel behavior analysis” and “traffic network analysis”. The seminar was held in the graduate course of Prof. Marco Nie on “Traffic Flow Theory”.]
Textbooks used:
- “Analysis and Modeling of Traffic Behavior: Theory/Model/Survey/Application” (Ryuichi Kitamura and Takayuki Morikawa [editors])
- “Equilibrium Analysis of Transportation Networks” (JSCE [editor])
Programming languages used:
- Python (some functions in Pythonbiogeme)
- R
【Introduction】
- 1: 01.pdf01code.zip
- 2: Data Aggregation and Visualization 02Py.pdf02R.pdf02code.zip
- 3: Statistical data analysis in R/Python 03Py.pdf03R.pdf03code.zip
【Travel Behavior Analsis Seminar】
- 4-5: Travel behavior theory & Travel behavior scheme 04_01.pdf
- 4-5: Travel Behavior Analysis & Econometric Approach to Behavior Analysis 04_02.pdf
- (P) Estimation of multinomial logit model (R, Python) 05Py.pdf05R.pdf05code.zip
- 6-7: Modeling: Discrete and continuous choice models, simultaneous equation models 06_01.pdf
- 6-7: Structural equation models, Dynamic models & Survival time models 06_02.pdf
- (P) Full information maximum likelihood method 07Py.pdf
- (P) Estimation of survival time model 07R.pdf07code.zip
- 8-9: Phenomenon analysis: Trip frequency, destination, transportation, route choice Activity-Based Approach 08_01.pdf
- 8-9: Phenomenon analysis: Car ownership Analysis of non-routine (holiday) traffic 08_02.pdf
- (P) Tourism Wi-Fi data analysis 09R_1.pdf09R_2.pdf09code.zip
【Traffic Network Analysis】
- 10-11: Chp.1-4 (Equilibrium problems and link cost functions) 10_01.pdf
- 10-11: Chp.1-4 (Choice Behavior & Behavior Analysis) 10_02.pdf
- (P) Shortest path algorithm 11Py.pdf11R.pdf11code.zip
- 12-13: User equilibrium model 12_01.pdf
- 12-13: Solving the user equilibrium model 12_02.pdf
- (P)User Equilibrium Algorithm (Frank Wolfe Algorithm) 13Py.pdf13R.pdf13code.zip
- 14-15: Stochastic user equilibrium model 14_01.pdf
- 14-15: Solving stochastic user equilibrium models 14_02.pdf
- (P) MSA & Dial Algorithm 15Py_1.pdf15Py_2.pdf15code.zip
