- Assessment 3 Select Any Appropriate Time Series Data The Data Must Be More Than 30 Observations Avoid Annual Data If P 1 (104.76 KiB) Viewed 37 times
- Assessment 3 Select Any Appropriate Time Series Data The Data Must Be More Than 30 Observations Avoid Annual Data If P 2 (27.24 KiB) Viewed 37 times
- Assessment 3 Select Any Appropriate Time Series Data The Data Must Be More Than 30 Observations Avoid Annual Data If P 3 (27.68 KiB) Viewed 37 times
ASSESSMENT 3 Select any appropriate time series data. The data must be more than 30 observations (avoid annual data if possible). This dataset will be modelled using univariate modelling and Box-Jenkins techniques. Collectively, study your lecture notes carefully, read Chapter 7 of the textbook in particular. Analyse the data sets according to the steps as given below. Univariate Modelling (Topic 2) Using appropriate statistical measurement and suitable graph, analyze your data and comment. Your comments should touch on their behavior pattern, encompassing the trend, cyclical, seasonal and irregular components. 1. Divide the time-series data into estimation (within-sample) and evaluation parts (out-of- sample) appropriately. State how much data are kept for fitting and for evaluation purposes. (Read Section 8.4). 2. Using univariate modelling techniques discussed in Topic 2, fit three (3) appropriate models to your data (within-sample). State the reason(s) for choosing the three models for your data. If you need to find the initial values' of the models, state the method you have used and the reason(s) for choosing that method. 3. For the within-sample data, present the summary of your model estimation and the error measures based on the three models (fitted). ARIMA Modelling (Topic 5) Model Identification 4. The next step is to evaluate your fitted models using out-of-sample data. This is done by comparing the error measures. From the three (3) models, you will obtain only one best forecast model; Model I 1. STA572 L July 2022 1. H. N. V. L 2. Model estimation and validation L. ii. Using the within-sample dataset, check for stationarity of the series by plotting the time plot, the ACF and the PACF. For the following steps you are to read and understand page 270; Section 7.6.4 of the textbook. You are also required to plot the ACFS and PACFs of the data in difference series to check for stationarity. Perform unit root tests to support your stationarity checking. If the data series is stationary, then based on the ACF and PACF, identify and write three (3) most appropriate ARIMA models which you think are appropriate for the data set. For seasonal data, you have to follow the steps discussed in the book. Estimate the three ARIMA models as identified in 1 (v) above. Perform statistical validation on the three ARIMA models and present the summary of the model estimation and validation. Next, you are to pick the best model; Model 2 out of the three (3) by evaluating their out-of-sample performance. This is done by generating forecast values on the evaluation part of the data set and calculating the error measures. Report the results and your conclusion.
Finally, which of the models (Model I and Model 2) do you think would perform the forecast best? Why? Hence, perform a one-step-head forecast - briefly explain the future forecast of your data. For this assessment, you are to report your analysis, results and conclusion in a report writing. All the works must involve all group members. No sleeping partners! You should explain all steps you have done, clearly. This is very important because clarity of the explanations will help you to get more marks. Graphs are important to support your arguments/explanations. Finally, kindly upload the report (one report/group) to Microsoft teams latest by July 16, 2022.
-~34 1 2 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 id A B Airline CO 2 US 3 AA 4 AA 5 AS 6 CO 7 DL 8 DL 9 DL 10 AA 11 CO 12 DL 13 DL 14 AA 15 AS 16 US 17 B6 18 CO 19 DL 20 CO 21 AS 22 DL 23 AS 24 HA 25 US 26 AS 27 B6 28 B6 29 HA 30 00 31 00 32 US 33 US 34 HA 35 00 36 9E 37 HA 38 00 70 ou C D E F G Flight AirportFre Airport To DayOfWe Time 3 269 SFO 1558 PHX 2400 LAX 2466 SFO 108 ANC 1094 LAX 1768 LAX 2722 PHX 2606 SFO 2538 LAS 223 ANC 1646 PHX 2055 SLC 2408 LAX 132 ANC 498 DEN 98 DEN 1496 LAS 1450 LAS 507 ONT 128 FAI 2223 ANC 112 ANC 17 LAS 122 ANC 114 ANC 766 BQN 768 PSE 206 HNL IAH CLT DFW DFW SEA IAH MSP DTW MSP ORD SEA ATL ATL DFW PDX CLT 4746 BIS 6466 MYK JFK IAH MSP IAH SEA SLC SEA HNL PHX SEA MCO MCO OGG MSP LAX 1011 EWR CLT 1983 BOS 108 HNL 6900 MKE 3975 GFK 102 HNL 4829 OMA 30 cen CLT KOA ORD MSP ITO MSP ATI wwwwwwwwwwwwwwwwwwwww 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 mmm 333 3 3 3 3 3 wwwwwwww 3 3 3 3 3 3 3 15 15 20 20 30 30 30 30 35 40 49 50 50 55 55 55 59 60 60 75 80 85 90 100 113 150 240 262 300 300 300 300 300 302 302 305 305 305 THE H Length 205 222 165 195 202 181 220 228 216 200 201 212 210 170 215 179 213 162 181 167 206 270 200 380 327 200 169 180 36 85 53 111 135 41 44 68 49 75 רח I Delay 1 1 1 1 0 1 0 0 1 1 1 1 0 0000000000 1 OOOOOOOOOOOO 0 0 0 1 0 0 0 0 0 0 0 0 0