Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: zz-x2580
EC902: Time Series - Revision lecture
Outline
• Exam structure
• General tips on test taking
• Discuss key concepts
• Take questions from the audience
• This lecture will be recorded in lecture capture
2 / 11
Exam Structure (1/2)
3 / 11
Exam structure (2/2)
Section A (Answer all questions)
3 questions × 10 marks each (Term 1)
Section B (Answer one of two)
1 question × 20 marks each (Term 1)
Section C (Answer all questions)
3 questions × 10 marks each (time series)
Section D (Answer one of two)
1 question × 20 marks each (time series)
4 / 11
General tips
• Attempt the section you are more comfortable first
• Questions will have sub-parts
• pay attention to how marks are allocated between sub-parts
• spend time and space wisely
• Number your answers correctly
• Even if you are unsure about the answer to a specific question, write
something generic - do not submit a blank answer!
• Sections C and D have options to choose from
• if you answer both the options, tell us which question you want to be
marked on.
• Read all your questions carefully. Your answer should address the
specific question.
5 / 11
Exam resources
• Departmental resources [here]
• AEP
• Mathtype/ Equation editor
• Typing Latex in Word
• Inserting images/ symbols
• Familiarise yourself with mitigating circumstances [link]
(e.g. being ill, fire alarm goes off, wifi crashes, who to contact, etc.)
Note:
• Do not copy and paste your notes! (plagiarism, academic
misconduct)
• rewrite definitions in your own words
6 / 11
Advice and Feedback - Term 3
• Monday, 11am to 1pm
• book an appointment [here]
• in-person meetings
• email me separately if you want to meet online or if these times do
not work for you.
7 / 11
Topics (1/2)
1. Univariate time series models
• Introduction: time series as a stochastic process, stationarity, weak
dependence, ARMA
• Definitions
• Stationary ARMA processes
• Theoretical properties: mean, variance, autocovariances and
autocorrelations
• Empirical modelling - Box and Jenkins methodology: identification,
estimation, diagnostic testing
• Forecasting with ARMA models: point forecast, forecast confidence
intervals
8 / 11
Topics (2/2)
2. Dynamic regression models with stationary variables
• Distributed lag (DL) models
• Autoregressive distributed lag (ADL) models
• Granger causality tests
• VAR models
3. Nonstationarity and cointegration
• Deterministic and stochastic trends (properties)
• Concept of integrated series: I(0), I(1), I(2), I(d)
• Testing for non-stationarity - DF, ADF
• Spurious regressions, cointegration and error correction models
4. VAR models (concept is sufficient - no derivations)
5. Analysis of panel data (concept is sufficient - no derivations)
