EEET2369 – Signals and Systems 1: Assignment 2 - Assignments should be submitted to Blackboard using the eReport Submission link no later than 11:59 pm on Friday 26th May 2017 (end of week 12). A 5% penalty per day late applies (including weekends). - Assignments MUST be submitted in either Word format (.doc or .docx) or portable document format (.pdf). Due to compatibility issues reports in other formats including (but not limited to) Open Office format (.odt) or Mac OS format (.pages) WILL NOT BE ACCEPTED! - The assignment is an individual task and is worth 10% of the overall course mark. - The assignment should be presented as follows: 1. Page 1: Cover Page (example): RMIT University School of Engineering EEET2369 – Signals and Systems 1 Assignment #2 Lecturer: Dr. Katrina Neville Student Name: …….. Student Number: ……… Submission Due Date: ……. 2. Page 2 onwards: 3. Answers to questions with appropriate predictions, calculations, MATLAB graphs and discussions. 4. References: Any sources used to find out more information. I.e. textbooks, journal/conference papers, websites, etc. must be in IEEE style format. 5. Appendix: MATLAB code and other results and calculations. The assignment will be assessed on the depth of conceptual understanding shown for each task. It is important not only to present correct results/ graphs/code but also to be able to analyse and discuss what your results are showing and how they link in with the concepts behind FIR and IIR filters.Assignment 2: FIR & IIR Filters in Stock Market Analysis This assignment is going to examine how digital filtering techniques can be applied to stock market analysis. To start with, a simple moving average (SMA) filter will be designed and applied to the closing prices of different companies (based on your student number). Secondly a more complex exponential moving average (EMA) filter will be developed and compared to the SMA. a) In stock market analysis the simple moving average (SMA) filter (a.k.a running average filter) is often used to smooth out stock prices, this effectively removes any ‘noise’ from the data and gives clearer indications of trend. The SMA filter can be applied to short term stock prices (i.e. over a few days or weeks), this will give an output with more fluctuations but one that will more closely reflect the trends in the stock market. Alternatively the SMA filter can be applied over a longer-term (i.e. many weeks or months) this gives a smoother output, but there will be delays in the filter ‘catching up’ with the trend when significant changes in market conditions occur. In MATLAB you will design an SMA filter with the following impulse response: [ ] n k L h n L k = ∑ − − = 1 0 1 [ ] δ You will then use this to filter the closing prices of stocks for the following companies (based on the 5th number in your student number): Student Number Company (and Excel file containing stock prices) 1 Seven West Media Ltd. (SevenWest.xlsx) 2 Crown Resorts Ltd. (Crown.xlsx) 3 Newcrest Mining Ltd. (Newcrest.xlsx) 4 Commonwealth Bank of Australia (CBA.xlsx) 5 Nine Entertainment Co. Holdings (Nine.xlsx) 6 National Australia Bank Ltd. (NAB.xlsx) 7 Lendlease Group (Lendlease.xlsx) 8 Westfield Corporation Ltd. (Westfield.xlsx) 9 BlueScope Steel Ltd. (Bluescope.xlsx) 0 Cochlear Ltd (Cochlear.xlsx) Table 1: Company and Excel file containing stock prices - For example if your student number was 3210987 you will analyse the closing prices of stocks for BlueScope Steel Ltd. - Manually work out (neatly typed) the general transfer function and difference equation for a running average filter of length (L). Also sketch the general impulse response of this filter. - In MATLAB create a short-term SMA filter (~5 days) and a longer-term SMA filter (~20 days) and apply these to the closing prices of your stocks (use MATLAB’s filter() function to do this). Plot the outputs in MATLAB.Discuss: 1. How well do these two filters smooth the stock price data? 2. Compare the delay in the two methods. For both filters, how many initial output samples/data points should be discarded before a reliable indication of trend can be obtained from this output? Why is this the case? 3. Use the MATLAB freqz() function on these two filters. Do they behave as low-pass, high-pass, band-pass or band-stop filters? How does the length of the filter affect the passband region (frequencies that the filter passes)? b) Another type of filter used in stock market analysis is the exponential moving average (EMA) filter, whereas the SMA filter behaved as an FIR filter the EMA filter is effectively an IIR filter with an impulse response of: h[n] = α ((1 − α )n u[n]) Where 1 2 + = L α and L is the length of time you’re interested in analysing (and how quickly the impulse response will exponentially decay). The EMA filter places more weight on the most recent stock prices, this allows the filter output to react quicker to changes in market trends. Again the EMA filter can be a short-term or long-term filter and is often used to indicate to traders the best times to buy and sell stocks; generally if a short-term average line crosses above a long-term average line this indicates a good time to buy stocks, when the short-term average line crosses below the long-term average line this indicates a good time to sell stocks. - Again manually work out (neatly typed) the general transfer function and difference equation for this EMA filter (you’ll need to perform a z-transform on the impulse response to obtain this information, check the section on ‘Inverse z-Transform’ in lecture 11 for information on how to do this). - In MATLAB create a short-term EMA filter (~5 days) and a longer-term EMA filter (~20 days) and again apply these to the closing prices of your stocks (use the filter’s transfer function (not impulse response) together with MATLAB’s filter() function to do this). Plot the outputs in MATLAB. Discuss: 1. Compare the behaviour of the EMA filters with the SMA filters from part a. 2. Using these long and short-term average lines, identify the times when it would have been good to buy and sell stocks in your company. Justify your answer from the explanation at: http://www.asx.com.au/prices/exponential_moving_average.htm 3. Again use the MATLAB freqz() function on these two filters and examine their frequency responses.Assessment Guide For Signals and Systems Assignment 0-49 Fail NN 50-59 Pass PA 60-69 Credit CR 70-79 Distinction DI 80-100 High Distinc’n HD Results (e.g. MATLAB figures/output) (30%) No results presented or there are fundamental flaws in the student’s understanding of the task and/or MATLAB code resulting in meaningless results. Some results are correct but many of the graphs contain errors resulting from poor understanding of the task and/or MATLAB syntax errors. Results are mostly correct. There may be a couple of graphs that contain minor errors and/or section contains superfluous or irrelevant results. Results presented are correct. Minimum required results presented to successfully answer assignment questions. All results presented in the report are correct and well presented. Student may also have included extra (relevant) results to help explain advanced concepts relating to the project. Discussion and analysis (50%) No relevant analysis has been presented in report. Student was unable to make links to theoretical concepts related to the topic and may have included irrelevant facts to explain results. Analysis presented in the report was superficial and only very basic links were made to the theoretical concepts related to the topic. Overall student appears to lack in-depth understanding. A reasonable analysis of the results has been presented, but it may lack some depth. Links have been made to theoretical concepts relating to the topic, but may lack essential details. A good analysis of results has been presented with only minor details missing. Student was able to make links to theoretical concepts relating to the topic to explain results. Student has presented an in-depth analysis of their results and has made links to advanced theoretical concepts relating to the topic to explain results. Referencing and citations (10%) Poor referencing style or no references used. Material may be copied without citing sources appropriately. References were inappropriate and/or lacked relevancy (e.g. Wikipedia or opinion pieces used). Citations may be missing. References are mostly appropriate and show some variation in type. A good attempt has been made at using the IEEE reference format with only minor errors. References used are all appropriate and show good variation in type. IEEE referencing format used correctly with very few errors. Has used a wide range of appropriate references. IEEE referencing format used flawlessly. Layout (10%) No structure or structure is highly disorganised. Poor use of grammar, and punctuation. Poor layout and difficult to read. Structure is sufficient to present the content. Ideas often presented in a disorganised manner. Grammar and spelling just acceptable. Content is generally organised logically with only some sections needing more attention. Few errors in spelling and grammar, report easy to navigate. Structure is sound throughout and follows a logical order. No errors in spelling and grammar, report clearly organised. Outstanding presentation of material which supports all requirements. Outstanding and professional use of language.