Strategy learner - Revision history

Ssung40: /* What to turn in */

2018-11-28T04:50:13Z

What to turn in

Ssung40: /* Rubric */

2018-11-28T04:40:05Z

Rubric

Ssung40: /* Rubric */

2018-11-28T04:36:50Z

Rubric

Ssung40: /* What to turn in */

2018-11-28T04:30:30Z

What to turn in

Ssung40: /* What to turn in */

2018-11-28T04:29:09Z

What to turn in

Ssung40 at 04:25, 28 November 2018

2018-11-28T04:25:15Z

Ssung40: /* Contents of Report */

2018-11-28T04:22:58Z

Contents of Report

Tucker: /* What to turn in */

2018-11-27T18:34:03Z

What to turn in

Tucker at 18:30, 27 November 2018

2018-11-27T18:30:28Z

Tucker at 18:22, 27 November 2018

2018-11-27T18:22:34Z

@@ Line 119: / Line 119: @@
 ** <tt>marketsimcode.py</tt> Optional if needed to support your strategy learner. An improved version of your marketsim code that accepts a "trades" data frame (instead of a file).
 ** <tt>StrategyLearner.py</tt>
-** <S><tt>ManualStrategy.py</tt></S>
+** <tt>ManualStrategy.py</tt>
 ** <tt>experiment1.py</tt>
 ** <tt>experiment2.py</tt>

@@ Line 181: / Line 181: @@
 ** When evaluating the trades generated by your learner, we '''will''' consider transaction costs (market  impact).
-<s>'''Report: 30 points'''
+'''Report: 30 points'''
-'''NEW''' Indicators (up to 15 points potential deductions):
+<s>'''NEW''' Indicators (up to 15 points potential deductions):
 * Is there a compelling description why each indicator might work (-2 for each, up to a total of 6 off)
 * Is each indicator described in sufficient detail that someone else could reproduce it? (-5 points for each if not)

@@ Line 181: / Line 181: @@
 ** When evaluating the trades generated by your learner, we '''will''' consider transaction costs (market  impact).
-'''Report: 30 points'''
+<s>'''Report: 30 points'''
 '''NEW''' Indicators (up to 15 points potential deductions):
@@ Line 188: / Line 188: @@
 * Is there a chart for each indicator that properly illustrates its operation? (-5 points for each if not)
 * Is at least one indicator different from those provided by the instructor's code (i.e., another indicator that is not SMA, Bollinger Bands or RSI) (-10 points if not)
 * Does the submitted code <tt>indicators.py</tt> properly reflect the indicators provided in the report (-20 points if not)

@@ Line 123: / Line 123: @@
 ** <tt>experiment2.py</tt>
 ** Your learning code: QLearner.py, or RTLearner and BagLearner.py or OptimizeLearner.py.
-* Make sure that you submit all code that is necessary for your software to run.  If your code crashes because of a missing file, you may lose signficant points on the code section.
+* Make sure that you submit all code that is necessary for your software to run including plotting.  If your code crashes because of a missing file, you may lose signficant points on the code section.
 * Do not submit any other files. Penalties will be assessed for additional files outside of the allowed ones as stated above.

@@ Line 124: / Line 124: @@
 ** Your learning code: QLearner.py, or RTLearner and BagLearner.py or OptimizeLearner.py.
 * Make sure that you submit all code that is necessary for your software to run.  If your code crashes because of a missing file, you may lose signficant points on the code section.
-* Do not submit any other files.
+* Do not submit any other files. Penalties will be assessed for additional files outside of the allowed ones as stated above.
 ==Rubric==

@@ Line 2: / Line 2: @@
 * It is OK to read data "before" the start date in order to ramp up indicators before start of trading date.  It is OK also NOT to do that, but in this case you can't start trading until after the indicators have enough lookback to work.
 * You should use the same indicators and indicator meta parameters as you used in manual strategy. One exception: If your learning method is setting the values of the meta parameters of the indicators, this is OK.
-* You should include at least 3 indicators in your learning strategy.
+* You should include at least 3 indicators in your learning strategy. *Caveat, if you used less than 3 indicators for manual strategy, please use the same for the machine learning strategy.
 * The author() method is required in each python file.
 * Plots are required to have proper plot components: labels on each axis, legend (if more than 1 variable), and title.  Points will deducted if they do not.

@@ Line 93: / Line 93: @@
 * Describe the steps you took to frame the trading problem as a learning problem for your learner. What are your indicators? Did you adjust the data in any way (dicretization, standardization)? Why or why not?
-* '''NEW''' Describe at least 3 and at most 5 technical indicators that you used.  You may find our lecture on time series processing to be helpful.  For each indicator you should create a single, compelling chart that illustrates the indicator.  As an example, you might create a chart that shows the price history of the stock, along with "helper data" (such as upper and lower bollinger bands) and the value of the indicator itself.  Another example: If you were using price/SMA as an indicator you would want to create a chart with 3 lines: Price, SMA, Price/SMA.  In order to facilitate visualization of the indicator you might normalize the data to 1.0 at the start of the date range (i.e. divide price[t] by price[0]). Your report description of each indicator should enable someone to reproduce it just by reading the description. We want a written description here, not code, however, it is OK to augment your written description with a pseudocode figure. At least one of the indicators you use should be completely different from the ones presented in our lectures. (i.e. something other than SMA, Bollinger Bands, RSI).
+<s>* '''NEW''' Describe at least 3 and at most 5 technical indicators that you used.  You may find our lecture on time series processing to be helpful.  For each indicator you should create a single, compelling chart that illustrates the indicator.  As an example, you might create a chart that shows the price history of the stock, along with "helper data" (such as upper and lower bollinger bands) and the value of the indicator itself.  Another example: If you were using price/SMA as an indicator you would want to create a chart with 3 lines: Price, SMA, Price/SMA.  In order to facilitate visualization of the indicator you might normalize the data to 1.0 at the start of the date range (i.e. divide price[t] by price[0]). Your report description of each indicator should enable someone to reproduce it just by reading the description. We want a written description here, not code, however, it is OK to augment your written description with a pseudocode figure. At least one of the indicators you use should be completely different from the ones presented in our lectures. (i.e. something other than SMA, Bollinger Bands, RSI).
 *Experiment 1: Using exactly the same indicators that you used in manual_strategy, compare your manual strategy with your learning strategy in sample.
 ** Describe your experiment in detail: Assumptions, parameter values and so on.
 ** Describe the outcome of your experiment.

@@ Line 1: / Line 1: @@
 '''Updated November 27, 2018'''
 * It is OK to read data "before" the start date in order to ramp up indicators before start of trading date.  It is OK also NOT to do that, but in this case you can't start trading until after the indicators have enough lookback to work.
-* o remove relation to manual strategy. Text that was removed from the earlier version of the assignment is indicated in <s>strike through</s>.
+* You should use the same indicators and indicator meta parameters as you used in manual strategy. One exception: If your learning method is setting the values of the meta parameters of the indicators, this is OK.
 '''Updated July 25, 2018''' to remove relation to manual strategy. Text that was removed from the earlier version of the assignment is indicated in <s>strike through</s>.

← Older revision		Revision as of 18:22, 27 November 2018
Line 1:		Line 1:
		+	'''Updated November 27, 2018'''
		+	* It is OK to read data "before" the start date in order to ramp up indicators before start of trading date. It is OK also NOT to do that, but in this case you can't start trading until after the indicators have enough lookback to work.
		+	* o remove relation to manual strategy. Text that was removed from the earlier version of the assignment is indicated in <s>strike through</s>.
		+
	'''Updated July 25, 2018''' to remove relation to manual strategy. Text that was removed from the earlier version of the assignment is indicated in <s>strike through</s>.		'''Updated July 25, 2018''' to remove relation to manual strategy. Text that was removed from the earlier version of the assignment is indicated in <s>strike through</s>.