Graphing - Revision history

JL at 12:04, 27 October 2015

2015-10-27T12:04:35Z

JL: /* Scatter Diagrams */

2015-10-27T11:57:04Z

‎Scatter Diagrams

JL: /* Line Plots */

2015-10-27T11:54:58Z

‎Line Plots

Rb: /* Observed Distributions/Kernel Density Plots */

2015-10-27T11:43:30Z

‎Observed Distributions/Kernel Density Plots

Rb: /* Theoretical Distributions */

2015-10-27T11:36:43Z

‎Theoretical Distributions

JL: /* Theoretical Distributions */

2015-10-25T19:47:04Z

‎Theoretical Distributions

JL: /* Theoretical Distributions */

2015-10-25T19:46:37Z

‎Theoretical Distributions

JL: /* Theoretical Distributions */

2015-10-25T19:44:56Z

‎Theoretical Distributions

JL at 19:41, 25 October 2015

2015-10-25T19:41:34Z

JL: /* Observed Distributions/Kernel Density Plots */

2015-10-25T19:38:26Z

‎Observed Distributions/Kernel Density Plots

← Older revision		Revision as of 12:04, 27 October 2015
Line 120:		Line 120:

	Here the <code>ksdensity</code> chose the values at which to evaluate the density itself. If you were planning to compare this to a different distribution, i.e. one of the theoretical distributions discussed above, you could also force the function to evaluate the density at the same x-coordinates by calling <code>ksdensity(stocks(:),x)</code> where <code>x</code> is defined as above.		Here the <code>ksdensity</code> chose the values at which to evaluate the density itself. If you were planning to compare this to a different distribution, i.e. one of the theoretical distributions discussed above, you could also force the function to evaluate the density at the same x-coordinates by calling <code>ksdensity(stocks(:),x)</code> where <code>x</code> is defined as above.
		+
		+	= Saving Plots =

@@ Line 44: / Line 44: @@
 In this example we use the wage data from the Wooldridge dataset 'wages1'
+[http://www.cengagebrain.com/cgi-wadsworth/course_products_wp.pl?fid=M20b&product_isbn_issn=9781111531041&token=8D04240DC39B22D05B49B265F2C8E62C6876DDE99FE979BC4A500075EC976963ED1045639B2C75C4B5B2337F07088998 Wooldridge data sets]
 <source>

@@ Line 24: / Line 24: @@
 [[File:lineplot1.png]]
-To modify an existing plot use the <code>hold on</code> command. Subsequent plots will then be added to the current figure. For example to add to a horizontal lines for the mean and another two indicating +/- 1 standard deviation from the mean for the stocks data:
+To modify an existing plot use the <code>hold all</code> command. Subsequent plots will then be added to the current figure. For example to add to a horizontal lines for the mean and another two indicating +/- 1 standard deviation from the mean for the stocks data:
 <source>
 avg=mean(stocks(:));
 sd=std(stocks(:));
-hold on
+hold all
 plot(xlim, [avg avg],'-k')
 plot(xlim, [avg+sd avg+sd],'--r')

@@ Line 109: / Line 109: @@
 == Observed Distributions/Kernel Density Plots ==
-Kernel Density estimation is a way to estimate the probability density function of a random variable from a sample of data. Returning to the stock market values data used above we can fit a PDF to this data using the <code>ksdensity</code> function as follows:
+Kernel Density estimation is a way to estimate the probability density function of a random variable from a sample of data. Returning to the stock market values data used above we can fit a PDF to this data using the <code>ksdensity</code> function. We input our observations and the function will return two outputs, <code>xi</code>, a vector of x-coordinates at which the density was evaluated and <code>fi</code> a corresponding vector of densities. We can then use these outputs to produce a nice plot as follows:
 <source>
@@ Line 118: / Line 118: @@
 </source>
 [[File:KernelPlot1.png]]

@@ Line 82: / Line 82: @@
 == Theoretical Distributions ==
-Matlab has a number of statistical distributions, see the [[StatFunct | entry on statistical distributions]] for an introduction. We can use the <code>plot</code> function to graph these distributions. The first thing to do is generate a vector of suitable values for the x-axis; in this example we use the <code>linspace</code> function to create a <math>(1000 \times 1)</math> vector of evenly spaced points on the x-axis. We then use this vector as the input to our distribution of choice (e.g. the Normal PDF) to obtain a vector of the probabilities for each corresponding x value; these will form our y-axis values in the plot.
+Matlab has a number of statistical distributions, see the [[StatFunct | entry on statistical distributions]] for an introduction. We can use the <code>plot</code> function to graph these distributions. The first thing to do is generate a vector of suitable values for the x-axis; in this example we use the <code>linspace</code> function to create a <math>(1000 \times 1)</math> vector of evenly spaced points between -5 and 5 on the x-axis. We then use this vector as the input to our distribution of choice (e.g. the Normal PDF) to obtain a vector of the probabilities for each corresponding x value; these will form our y-axis values in the plot.
 <source>

@@ Line 15: / Line 15: @@
 </source>
-The data contains a (100x10) matrix of stock market observations and we are going to plot all of these data point on a single line plot using the <source enclose="none">plot</source> function.
+The data contains a (100x10) matrix of stock market observations and we are going to plot all of these data point on a single line plot using the <code>plot</code> function.
 <source>
@@ Line 24: / Line 24: @@
 [[File:lineplot1.png]]
-To modify an existing plot use the <source enclose="none">hold on</source> command. Subsequent plots will then be added to the current figure. For example to add to a horizontal lines for the mean and another two indicating +/- 1 standard deviation from the mean for the stocks data:
+To modify an existing plot use the <code>hold on</code> command. Subsequent plots will then be added to the current figure. For example to add to a horizontal lines for the mean and another two indicating +/- 1 standard deviation from the mean for the stocks data:
 <source>
@@ Line 37: / Line 37: @@
 [[File:lineplot2.png]]
-Here, in each case the <source enclose="none">plot</source> function is plotting a straight line between two points. The first argument, <source enclose="none">xlim</source> queries the figure to get the coordinates of the start and end of the x-axis (a two element vector), provides the x coordinates. The second argument is a vector containing the corresponding y coordinates of the two points.
+Here, in each case the <code>plot</code> function is plotting a straight line between two points. The first argument, <code>xlim</code> queries the figure to get the coordinates of the start and end of the x-axis (a two element vector), provides the x coordinates. The second argument is a vector containing the corresponding y coordinates of the two points.
 The '-k' argument specifies the plotted line to be a solid black line, whereas the '--r' arguments specifies the plotted line as a red dashed line. Matlab supports a number of line (and marker) types of varying colours; for full details refer to Matlab's documentation on [http://uk.mathworks.com/help/matlab/ref/linespec.html LineSpec].

← Older revision		Revision as of 19:38, 25 October 2015
Line 109:		Line 109:
	== Observed Distributions/Kernel Density Plots ==		== Observed Distributions/Kernel Density Plots ==

		+	Kernel Density estimation is a way to estimate the probability density function of a random variable from a sample of data. Returning to the stock market values data used above we can fit a PDF to this data using the <code>ksdensity</code> function as follows:

	<source>		<source>