Navigation

← Previous Changeset
Next Changeset →

Changeset 173

Timestamp:

05/22/08 09:30:08 (6 months ago)

Author:

snovik

Message:

Fix: Risk & Gaussian Statistics
New: RiskStat? test

Files:

trunk/QLNet/QLNet.sln (modified) (2 diffs)
trunk/QLNet/QLNet.vsmdi (modified) (3 diffs)
trunk/QLNet/QLNet/Math/randomnumbers/sobolrsg.cs (modified) (1 diff)
trunk/QLNet/QLNet/Math/statistics/gaussianstatistics.cs (modified) (2 diffs)
trunk/QLNet/QLNet/Math/statistics/generalstatistics.cs (modified) (12 diffs)
trunk/QLNet/QLNet/Math/statistics/incrementalstatistics.cs (modified) (2 diffs)
trunk/QLNet/QLNet/Math/statistics/riskstatistics.cs (modified) (4 diffs)
trunk/QLNet/QLNet/Types.cs (modified) (1 diff)
trunk/QLNet/Test2008/T_RiskStats.cs (modified) (2 diffs)

Legend:

: Unmodified
: Added
: Removed
: Modified
: Copied
: Moved

trunk/QLNet/QLNet.sln

r171	r173
2	2	Microsoft Visual Studio Solution File, Format Version 10.00
3	3	# Visual Studio 2008
4		~~Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "Solution Items", "Solution Items", "{A1B2F21C-BD08-4751-9FBE-6DD0405E3DBC}"~~
5		~~ProjectSection(SolutionItems) = preProject~~
6		~~localtestrun.testrunconfig = localtestrun.testrunconfig~~
7		~~QLNet.vsmdi = QLNet.vsmdi~~
8		~~QLNet1.vsmdi = QLNet1.vsmdi~~
9		~~EndProjectSection~~
10		~~EndProject~~
11	4	Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "QLNet", "QLNet\QLNet.csproj", "{F6E762BD-DCDF-4CA0-ABAD-CB21C7D03BEC}"
12	5	EndProject
…	…
27	20	Project("{F184B08F-C81C-45F6-A57F-5ABD9991F28F}") = "Swap", "VB\Swap\Swap.vbproj", "{5724FF23-DB57-4D10-B88E-B1DFF1249C88}"
28	21	EndProject
	22	Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "Solution Items", "Solution Items", "{0A13061B-C0D9-4D34-981A-E506240D6A75}"
	23	ProjectSection(SolutionItems) = preProject
	24	QLNet.vsmdi = QLNet.vsmdi
	25	EndProjectSection
	26	EndProject
29	27	Global
30	28	GlobalSection(TestCaseManagementSettings) = postSolution
31		CategoryFile = QLNet1.vsmdi
	29	CategoryFile = QLNet.vsmdi
32	30	EndGlobalSection
33	31	GlobalSection(SolutionConfigurationPlatforms) = preSolution

trunk/QLNet/QLNet.vsmdi

r165	r173
5	5	<TestLink id="bbe2d937-bb70-a666-a879-b3063800a42e" name="testFairSpread" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
6	6	<TestLink id="25b4118b-8066-f12f-3687-a5ef1886eb1c" name="testInArrears" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
	7	<TestLink id="0ee842b0-6275-252f-a1d0-000b0a6a4c52" name="testCachedValue" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
7	8	<TestLink id="27ded393-6046-4057-bd65-31f629c9bf55" name="testFairRate" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
8		<TestLink id="0ee842b0-6275-252f-a1d0-000b0a6a4c52" name="testCachedValue" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
9	9	<TestLink id="3b1816a9-7785-f64f-bc52-7014af11e8e5" name="testSpreadDependency" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
10	10	<TestLink id="2b04fe68-610a-b2de-e241-ef9a50fdb3fb" name="testRateDependency" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
…	…
42	42	</TestLinks>
43	43	</TestList>
	44	<TestList name="Risk Stats" id="a33eb53e-6299-480c-8958-5bcb922a4801" parentListId="8c43106b-9dc1-4907-a29f-aa66a61bf5b6">
	45	<TestLinks>
	46	<TestLink id="b87fb037-4727-72de-6cc9-162a99e11210" name="RiskStatisticsTest" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
	47	</TestLinks>
	48	</TestList>
44	49	<TestList name="SampledCurve" id="a6f5cea2-0706-4e42-a89c-afd9f63d2645" parentListId="8c43106b-9dc1-4907-a29f-aa66a61bf5b6">
45	50	<TestLinks>
…	…
49	54	<TestList name="American Option" id="da9a808d-acf7-4baf-9dfb-e4b091d30fef" parentListId="8c43106b-9dc1-4907-a29f-aa66a61bf5b6">
50	55	<TestLinks>
51		<TestLink id="~~33c4d801-48ac-44c4-329e-d6fe84f5e034" name="testJu~~Values" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
	56	<TestLink id="2bbf531b-68cb-84c0-a06f-6bedfb89f20f" name="testBaroneAdesiWhaleyValues" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
52	57	<TestLink id="3be50179-04ee-902d-73d8-3c94e9305a67" name="testFdShoutGreeks" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
53	58	<TestLink id="783f96db-cce8-bf61-b280-f5010f7d7eb3" name="testFdValues" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
54	59	<TestLink id="1e42c90a-8e82-1369-9ed9-52b036b1c07e" name="testBjerksundStenslandValues" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
55	60	<TestLink id="9f4ff6f8-9938-8ecb-5608-7dbc928cbda3" name="testFdAmericanGreeks" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
56		<TestLink id="~~2bbf531b-68cb-84c0-a06f-6bedfb89f20f" name="testBaroneAdesiWhaley~~Values" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
	61	<TestLink id="33c4d801-48ac-44c4-329e-d6fe84f5e034" name="testJuValues" storage="test2008\bin\debug\test2008.dll" type="Microsoft.VisualStudio.TestTools.TestTypes.Unit.UnitTestElement, Microsoft.VisualStudio.QualityTools.Tips.UnitTest.ObjectModel, PublicKeyToken=b03f5f7f11d50a3a" />
57	62	</TestLinks>
58	63	</TestList>

trunk/QLNet/QLNet/Math/randomnumbers/sobolrsg.cs

r172	r173
116	116	/! \pre dimensionality must be <= PPMT_MAX_DIM /
117	117	//public SobolRsg(int dimensionality, ulong seed = 0, DirectionIntegers directionIntegers = Jaeckel);
	118	public SobolRsg(int dimensionality) : this(dimensionality, 0, DirectionIntegers.Jaeckel) { }
118	119	public SobolRsg(int dimensionality, ulong seed, DirectionIntegers directionIntegers) {
119	120	dimensionality_ = dimensionality;

trunk/QLNet/QLNet/Math/statistics/gaussianstatistics.cs

r172	r173
33	33
34	34	public GenericGaussianStatistics() { }
35		// public GenericGaussianStatistics(Stat s) : base(s) { }
	35	public GenericGaussianStatistics(Stat s) {
	36	impl_ = s;
	37	}
36	38
37	39	#region wrap-up Stat
38		Stat impl_ = new Stat();
	40	protected Stat impl_ = new Stat();
39	41
40	42	public int samples() { return impl_.samples(); }
41	43	public double mean() { return impl_.mean(); }
	44	public double min() { return impl_.min(); }
	45	public double max() { return impl_.max(); }
42	46	public double standardDeviation() { return impl_.standardDeviation(); }
	47	public double variance() { return impl_.variance(); }
	48	public double skewness() { return impl_.skewness(); }
	49	public double kurtosis() { return impl_.kurtosis(); }
43	50	public double percentile(double percent) { return impl_.percentile(percent); }
	51	public double weightSum() { return impl_.weightSum(); }
	52
	53	public void reset() { impl_.reset(); }
	54	public void addSequence(List<double> data, List<double> weight) { impl_.addSequence(data, weight); }
44	55
45	56	public KeyValuePair<double, int> expectationValue(Func<KeyValuePair<double, double>, double> f,
…	…
168	179	public class GaussianStatistics : GenericGaussianStatistics<GeneralStatistics> { }
169	180
	181
	182	//! Helper class for precomputed distributions
	183	public class StatsHolder : IGeneralStatistics {
	184	private double mean_, standardDeviation_;
	185	public double mean() { return mean_; }
	186	public double standardDeviation() { return standardDeviation_; }
	187
	188	public StatsHolder() { } // required for generics
	189	public StatsHolder(double mean, double standardDeviation) {
	190	mean_ = mean;
	191	standardDeviation_ = standardDeviation;
	192	}
	193
	194	#region IGeneralStatistics
	195	public int samples() { throw new NotSupportedException(); }
	196	public double min() { throw new NotSupportedException(); }
	197	public double max() { throw new NotSupportedException(); }
	198	public double variance() { throw new NotSupportedException(); }
	199	public double skewness() { throw new NotSupportedException(); }
	200	public double kurtosis() { throw new NotSupportedException(); }
	201	public double percentile(double percent) { throw new NotSupportedException(); }
	202	public double weightSum() { throw new NotSupportedException(); }
	203
	204	public void reset() { throw new NotSupportedException(); }
	205	public void addSequence(List<double> data, List<double> weight) { throw new NotSupportedException(); }
	206
	207	public KeyValuePair<double, int> expectationValue(Func<KeyValuePair<double, double>, double> f,
	208	Func<KeyValuePair<double, double>, bool> inRange) {
	209	throw new NotSupportedException();
	210	}
	211	#endregion
	212	}
170	213	}

trunk/QLNet/QLNet/Math/statistics/generalstatistics.cs

r172	r173
26	26	int samples();
27	27	double mean();
	28	double min();
	29	double max();
28	30	double standardDeviation();
	31	double variance();
	32	double skewness();
	33	double kurtosis();
29	34	double percentile(double percent);
	35	double weightSum();
	36
	37	void reset();
	38	void addSequence(List<double> data, List<double> weight);
30	39
31	40	KeyValuePair<double, int> expectationValue(Func<KeyValuePair<double, double>, double> f,
…	…
51	60
52	61	private bool sorted_;
	62	private double? mean_ = null, weightSum_ = null, variance_ = null, skewness_ = null, kurtosis_ = null;
53	63
54	64
…	…
74	84
75	85	//! adds a datum to the set, possibly with a weight
76		~~//public void add(double value, double weight = 1.0) {~~
	86	public void add(double value) { add(value, 1); }
77	87	public void add(double value, double weight) {
78	88	if (!(weight>=0.0)) throw new ApplicationException("negative weight not allowed");
79	89	samples_.Add(new KeyValuePair<double,double>(value,weight));
	90
80	91	sorted_ = false;
	92	mean_ = weightSum_ = variance_ = skewness_ = kurtosis_ = null;
81	93	}
82	94
…	…
84	96	public void reset(){
85	97	samples_ = new List<KeyValuePair<double,double>>();
	98
86	99	sorted_ = true;
	100	mean_ = weightSum_ = variance_ = skewness_ = kurtosis_ = null;
87	101	}
88	102
…	…
90	104	public void sort() {
91	105	if (!sorted_) {
92		samples_.Sort();
	106	samples_.Sort((x, y) => x.Key.CompareTo(y.Key));
93	107	sorted_ = true;
94	108	}
…	…
98	112	//! sum of data weights
99	113	public double weightSum() {
100		~~double result = 0.0;~~
101		~~result~~ = samples_.Sum<KeyValuePair<double, double>>(x => x.Value);
102		return ~~result~~;
	114	if (weightSum_ == null)
	115	weightSum_ = samples_.Sum<KeyValuePair<double, double>>(x => x.Value);
	116	return weightSum_.GetValueOrDefault();
103	117	}
104	118
…	…
106	120	\f[ \langle x \rangle = \frac{\sum w_i x_i}{\sum w_i}. \f] */
107	121	public double mean() {
108		int N = samples();
109		if (!(samples() > 0)) throw new ApplicationException("empty sample set");
110		// eat our own dog food
111		return expectationValue(x => x.Key * x.Value, x => true).Key;
	122	if (mean_ == null) {
	123	int N = samples();
	124	if (!(samples() > 0)) throw new ApplicationException("empty sample set");
	125	// eat our own dog food
	126	mean_ = expectationValue(x => x.Key * x.Value, x => true).Key;
	127	}
	128	return mean_.GetValueOrDefault();
112	129	}
113	130
…	…
120	137	x-\langle x \rangle \right)^2 \right\rangle. \f] */
121	138	public double variance() {
122		int N = samples();
123		if (!(N > 1)) throw new ApplicationException("sample number <=1, unsufficient");
124		// Subtract the mean and square. Repeat on the whole range.
125		// Hopefully, the whole thing will be inlined in a single loop.
126		double s2 = expectationValue(x => Math.Pow(x.Key * x.Value - mean(), 2), x => true).Key;
	139	if (variance_ == null) {
	140	int N = samples();
	141	if (!(N > 1)) throw new ApplicationException("sample number <=1, unsufficient");
	142	// Subtract the mean and square. Repeat on the whole range.
	143	// Hopefully, the whole thing will be inlined in a single loop.
	144	double s2 = expectationValue(x => Math.Pow(x.Key * x.Value - mean(), 2), x => true).Key;
127	145
128	146	//compose(square<Real>(), std::bind2nd(std::minus<Real>(), mean())), () => true).Key;
129		return s2*N/(N-1.0);
	147	variance_ = s2 * N / (N - 1.0);
	148	}
	149	return variance_.GetValueOrDefault();
130	150	}
131	151
…	…
136	156	*/
137	157	public double skewness() {
138		int N = samples();
139		if (!(N > 2)) throw new ApplicationException("sample number <=2, unsufficient");
140
141		double x = expectationValue(y => Math.Pow(y.Key * y.Value - mean(), 3), y => true).Key;
142		double sigma = standardDeviation();
143
144		return (x/(sigmasigmasigma))(N/(N-1.0))(N/(N-2.0));
	158	if (skewness_ == null) {
	159	int N = samples();
	160	if (!(N > 2)) throw new ApplicationException("sample number <=2, unsufficient");
	161
	162	double x = expectationValue(y => Math.Pow(y.Key * y.Value - mean(), 3), y => true).Key;
	163	double sigma = standardDeviation();
	164
	165	skewness_ = (x / Math.Pow(sigma, 3)) * (N / (N - 1.0)) * (N / (N - 2.0));
	166	}
	167	return skewness_.GetValueOrDefault();
145	168	}
146	169
…	…
152	175	*/
153	176	public double kurtosis() {
154		int N = samples();
155		if (!(N > 3)) throw new ApplicationException("sample number <=3, unsufficient");
156
157		double x = expectationValue(y => Math.Pow(y.Key * y.Value - mean(), 4), y => true).Key;
158		double sigma2 = variance();
159
160		double c1 = (N/(N-1.0)) * (N/(N-2.0)) * ((N+1.0)/(N-3.0));
161		double c2 = 3.0 * ((N-1.0)/(N-2.0)) * ((N-1.0)/(N-3.0));
162
163		return c1(x/(sigma2sigma2))-c2;
	177	if (kurtosis_ == null) {
	178	int N = samples();
	179	if (!(N > 3)) throw new ApplicationException("sample number <=3, unsufficient");
	180
	181	double x = expectationValue(y => Math.Pow(y.Key * y.Value - mean(), 4), y => true).Key;
	182	double sigma2 = variance();
	183
	184	double c1 = (N / (N - 1.0)) * (N / (N - 2.0)) * ((N + 1.0) / (N - 3.0));
	185	double c2 = 3.0 * ((N - 1.0) / (N - 2.0)) * ((N - 1.0) / (N - 3.0));
	186
	187	kurtosis_ = c1 * (x / (sigma2 * sigma2)) - c2;
	188	}
	189	return kurtosis_.GetValueOrDefault();
164	190	}
165	191
…	…
178	204	int N = 0;
179	205
180		foreach(KeyValuePair<double,double> x in samples_) {
181		if (inRange(x)) {
182		num += f(x)*x.Value;
183		den += x.Value;
184		N += 1;
185		}
	206	foreach(KeyValuePair<double,double> x in samples_.Where<KeyValuePair<double,double>>(x => inRange(x))) {
	207	num += f(x)*x.Value;
	208	den += x.Value;
	209	N += 1;
186	210	}
187	211
…	…
233	257	}
234	258
235		////! adds a sequence of data to the set, with default weight
236		//template <class DataIterator>
237		//void addSequence(DataIterator begin, DataIterator end) {
238		// for (;begin!=end;++begin)
239		// add(*begin);
240		//}
241		////! adds a sequence of data to the set, each with its weight
242		//template <class DataIterator, class WeightIterator>
243		//void addSequence(DataIterator begin, DataIterator end,
244		// WeightIterator wbegin) {
245		// for (;begin!=end;++begin,++wbegin)
246		// add(begin, wbegin);
247		//}
	259	//! adds a sequence of data to the set, with default weight
	260	public void addSequence(List<double> list) {
	261	foreach(double v in list)
	262	add(v);
	263	}
	264	//! adds a sequence of data to the set, each with its weight
	265	public void addSequence(List<double> data, List<double> weight) {
	266	for(int i=0; i< data.Count; i++)
	267	add(data[i], weight[i]);
	268	}
248	269	}
249	270	}

trunk/QLNet/QLNet/Math/statistics/incrementalstatistics.cs

r172	r173
177	177	//! adds a datum to the set, possibly with a weight
178	178	/! \pre weight must be positive or null /
179		~~//void add(Real value, Real weight = 1.0);~~
	179	public void add(double value) { add(value, 1); }
180	180	public void add(double value, double weight) {
181	181	if (!(weight>=0.0)) throw new ApplicationException("negative weight (" + weight + ") not allowed");
…	…
223	223	}
224	224
225		////! adds a sequence of data to the set, with default weight
226		//template <class DataIterator>
227		//void addSequence(DataIterator begin, DataIterator end) {
228		// for (;begin!=end;++begin)
229		// add(*begin);
230		//}
231		////! adds a sequence of data to the set, each with its weight
232		///! \pre weights must be positive or null /
233		//template <class DataIterator, class WeightIterator>
234		//void addSequence(DataIterator begin, DataIterator end,
235		// WeightIterator wbegin) {
236		// for (;begin!=end;++begin,++wbegin)
237		// add(begin, wbegin);
238		//}
	225	//! adds a sequence of data to the set, with default weight
	226	public void addSequence(List<double> list) {
	227	foreach (double v in list)
	228	add(v, 1);
	229	}
	230	//! adds a sequence of data to the set, each with its weight
	231	/! \pre weights must be positive or null /
	232	public void addSequence(List<double> data, List<double> weight) {
	233	for (int i = 0; i < data.Count; i++)
	234	add(data[i], weight[i]);
	235	}
239	236	}
240	237	}

trunk/QLNet/QLNet/Math/statistics/riskstatistics.cs

r172	r173
35	35
36	36	#region wrap-up Stat
37		Stat impl_ = new Stat();
	37	protected Stat impl_ = new Stat();
38	38
39	39	public int samples() { return impl_.samples(); }
40	40	public double mean() { return impl_.mean(); }
	41	public double min() { return impl_.min(); }
	42	public double max() { return impl_.max(); }
41	43	public double standardDeviation() { return impl_.standardDeviation(); }
	44	public double variance() { return impl_.variance(); }
	45	public double skewness() { return impl_.skewness(); }
	46	public double kurtosis() { return impl_.kurtosis(); }
42	47	public double percentile(double percent) { return impl_.percentile(percent); }
	48	public double weightSum() { return impl_.weightSum(); }
	49
	50	public void reset() { impl_.reset(); }
	51	public void addSequence(List<double> data, List<double> weight) { impl_.addSequence(data, weight); }
43	52
44	53	public KeyValuePair<double, int> expectationValue(Func<KeyValuePair<double, double>, double> f,
…	…
79	88	public double regret(double target) {
80	89	// average over the range below the target
81		KeyValuePair<double, int> result = expectationValue(z => Math.Pow(z.Key * z.Value - mean(), 2),
82		z => z.~~Value~~ < target);
	90	KeyValuePair<double, int> result = expectationValue(z => Math.Pow(z.Key - target, 2),
	91	z => z.Key < target);
83	92	double x = result.Key;
84	93	int N = result.Value;
…	…
153	162	*/
154	163	public double averageShortfall(double target) {
155		KeyValuePair<double, int> result = expectationValue(z => ~~z.Key - target~~, z => z.Key < target);
	164	KeyValuePair<double, int> result = expectationValue(z => target - z.Key, z => z.Key < target);
156	165	double x = result.Key;
157	166	int N = result.Value;
…	…
164	173	/! \test the correctness of the returned values is tested by checking them against numerical calculations. /
165	174	//typedef GenericRiskStatistics<GaussianStatistics> RiskStatistics;
166		public class RiskStatistics : GenericRiskStatistics<GaussianStatistics> { }
	175	public class RiskStatistics : GenericRiskStatistics<GaussianStatistics> {
	176	public double gaussianPercentile(double value) {
	177	return ((GaussianStatistics)impl_).gaussianPercentile(value);
	178	}
	179	public double gaussianPotentialUpside(double value) {
	180	return ((GaussianStatistics)impl_).gaussianPotentialUpside(value);
	181	}
	182	public double gaussianValueAtRisk(double value) {
	183	return ((GaussianStatistics)impl_).gaussianValueAtRisk(value);
	184	}
	185	public double gaussianExpectedShortfall(double value) {
	186	return ((GaussianStatistics)impl_).gaussianExpectedShortfall(value);
	187	}
	188	public double gaussianShortfall(double value) {
	189	return ((GaussianStatistics)impl_).gaussianShortfall(value);
	190	}
	191	public double gaussianAverageShortfall(double value) {
	192	return ((GaussianStatistics)impl_).gaussianAverageShortfall(value);
	193	}
	194	public double gaussianRegret(double value) {
	195	return ((GaussianStatistics)impl_).gaussianRegret(value);
	196	}
	197	public double gaussianDownsideVariance() {
	198	return ((GaussianStatistics)impl_).gaussianDownsideVariance();
	199	}
	200	}
167	201	}

trunk/QLNet/QLNet/Types.cs

r172	r173
28	28
29	29	public const double M_LN2 = 0.693147180559945309417;
	30	public const double M_PI = 3.141592653589793238462643383280;
30	31	}
31	32

trunk/QLNet/Test2008/T_RiskStats.cs

r172	r173
25	25
26	26	namespace TestSuite {
27		class IncrementalGaussianStatistics : GenericGaussianStatistics<IncrementalStatistics> { }
	27	class IncrementalGaussianStatistics : GenericGaussianStatistics<IncrementalStatistics> {
	28	public double downsideVariance() { return ((IncrementalStatistics)impl_).downsideVariance(); }
	29	}
28	30
29	31	[TestClass()]
30		class T_RiskStats {
	32	public class T_RiskStats {
31	33	[TestMethod()]
32	34	public void RiskStatisticsTest() {
…	…
50	52	InverseCumulativeNormal inverseCum = new InverseCumulativeNormal(averages[i],sigmas[j]);
51	53
52		// SobolRsg rng(1);
53		// dataMin = QL_MAX_REAL;
54		// dataMax = QL_MIN_REAL;
55		// for (k=0; k<N; k++) {
56		// data[k] = inverseCum(rng.nextSequence().value[0]);
57		// dataMin = std::min(dataMin, data[k]);
58		// dataMax = std::max(dataMax, data[k]);
59		// weights[k]=1.0;
60		// }
61
62		// igs.addSequence(data.begin(),data.end(),weights.begin());
63		// s.addSequence(data.begin(),data.end(),weights.begin());
64
65		// // checks
66		// Real calculated, expected;
67		// Real tolerance;
68
69		// if (igs.samples() != N)
70		// BOOST_FAIL("IncrementalGaussianStatistics: "
71		// << "wrong number of samples\n"
72		// << " calculated: " << igs.samples() << "\n"
73		// << " expected: " << N);
74		// if (s.samples() != N)
75		// BOOST_FAIL("RiskStatistics: wrong number of samples\n"
76		// << " calculated: " << s.samples() << "\n"
77		// << " expected: " << N);
78
79
80		// // weightSum()
81		// tolerance = 1e-10;
82		// expected = std::accumulate(weights.begin(),weights.end(),0.0);
83		// calculated = igs.weightSum();
84		// if (std::fabs(calculated-expected) > tolerance)
85		// BOOST_FAIL("IncrementalGaussianStatistics: "
86		// << "wrong sum of weights\n"
87		// << std::setprecision(16)
88		// << " calculated: " << calculated << "\n"
89		// << " expected: " << expected << "\n"
90		// << " tolerance: " << tolerance);
91		// calculated = s.weightSum();
92		// if (std::fabs(calculated-expected) > tolerance)
93		// BOOST_FAIL("RiskStatistics: wrong sum of weights\n"
94		// << std::setprecision(16)
95		// << " calculated: " << calculated << "\n"
96		// << " expected: " << expected << "\n"
97		// << " tolerance: " << tolerance);
98
99
100		// // min
101		// tolerance = 1e-12;
102		// expected = dataMin;
103		// calculated = igs.min();
104		// if (std::fabs(calculated-expected)>tolerance)
105		// BOOST_FAIL("IncrementalGaussianStatistics: "
106		// << "wrong minimum value\n"
107		// << std::setprecision(16)
108		// << " calculated: " << calculated << "\n"
109		// << " expected: " << expected << "\n"
110		// << " tolerance: " << tolerance);
111		// calculated = s.min();
112		// if (std::fabs(calculated-expected)>tolerance)
113		// BOOST_FAIL("RiskStatistics: "
114		// << "wrong minimum value\n"
115		// << std::setprecision(16)
116		// << " calculated: " << calculated << "\n"
117		// << " expected: " << expected << "\n"
118		// << " tolerance: " << tolerance);
119
120
121		// // max
122		// expected = dataMax;
123		// calculated = igs.max();
124		// if (std::fabs(calculated-expected)>tolerance)
125		// BOOST_FAIL("IncrementalGaussianStatistics: "
126		// << "wrong maximum value\n"
127		// << std::setprecision(16)
128		// << " calculated: " << calculated << "\n"
129		// << " expected: " << expected << "\n"
130		// << " tolerance: " << tolerance);
131		// calculated = s.max();
132		// if (std::fabs(calculated-expected)>tolerance)
133		// BOOST_FAIL("RiskStatistics: "
134		// << "wrong maximum value\n"
135		// << std::setprecision(16)
136		// << " calculated: " << calculated << "\n"
137		// << " expected: " << expected << "\n"
138		// << " tolerance: " << tolerance);
139
140
141		// // mean
142		// expected = averages[i];
143		// tolerance = (expected == 0.0 ? 1.0e-13 :
144		// std::fabs(expected)*1.0e-13);
145		// calculated = igs.mean();
146		// if (std::fabs(calculated-expected) > tolerance)
147		// BOOST_FAIL("IncrementalGaussianStatistics: "
148		// << "wrong mean value"
149		// << " for N(" << averages[i] << ", "
150		// << sigmas[j] << ")\n"
151		// << std::setprecision(16)
152		// << " calculated: " << calculated << "\n"
153		// << " expected: " << expected << "\n"
154		// << " tolerance: " << tolerance);
155		// calculated = s.mean();
156		// if (std::fabs(calculated-expected) > tolerance)
157		// BOOST_FAIL("RiskStatistics: wrong mean value"
158		// << " for N(" << averages[i] << ", "
159		// << sigmas[j] << ")\n"
160		// << std::setprecision(16)
161		// << " calculated: " << calculated << "\n"
162		// << " expected: " << expected << "\n"
163		// << " tolerance: " << tolerance);
164
165
166		// // variance
167		// expected = sigmas[j]*sigmas[j];
168		// tolerance = expected*1.0e-1;
169		// calculated = igs.variance();
170		// if (std::fabs(calculated-expected) > tolerance)
171		// BOOST_FAIL("IncrementalGaussianStatistics: "
172		// << "wrong variance"
173		// << " for N(" << averages[i] << ", "
174		// << sigmas[j] << ")\n"
175		// << std::setprecision(16)
176		// << " calculated: " << calculated << "\n"
177		// << " expected: " << expected << "\n"
178		// << " tolerance: " << tolerance);
179		// calculated = s.variance();
180		// if (std::fabs(calculated-expected) > tolerance)
181		// BOOST_FAIL("RiskStatistics: wrong variance"
182		// << " for N(" << averages[i] << ", "
183		// << sigmas[j] << ")\n"
184		// << std::setprecision(16)
185		// << " calculated: " << calculated << "\n"
186		// << " expected: " << expected << "\n"
187		// << " tolerance: " << tolerance);
188
189
190		// // standardDeviation
191		// expected = sigmas[j];
192		// tolerance = expected*1.0e-1;
193		// calculated = igs.standardDeviation();
194		// if (std::fabs(calculated-expected) > tolerance)
195		// BOOST_FAIL("IncrementalGaussianStatistics: "
196		// << "wrong standard deviation"
197		// << " for N(" << averages[i] << ", "
198		// << sigmas[j] << ")\n"
199		// << std::setprecision(16)
200		// << " calculated: " << calculated << "\n"
201		// << " expected: " << expected << "\n"
202		// << " tolerance: " << tolerance);
203		// calculated = s.standardDeviation();
204		// if (std::fabs(calculated-expected) > tolerance)
205		// BOOST_FAIL("RiskStatistics: wrong standard deviation"
206		// << " for N(" << averages[i] << ", "
207		// << sigmas[j] << ")\n"
208		// << std::setprecision(16)
209		// << " calculated: " << calculated << "\n"
210		// << " expected: " << expected << "\n"
211		// << " tolerance: " << tolerance);
212
213
214		// // missing errorEstimate() test
215
216		// // skewness
217		// expected = 0.0;
218		// tolerance = 1.0e-4;
219		// calculated = igs.skewness();
220		// if (std::fabs(calculated-expected) > tolerance)
221		// BOOST_FAIL("IncrementalGaussianStatistics: "
222		// << "wrong skewness"
223		// << " for N(" << averages[i] << ", "
224		// << sigmas[j] << ")\n"
225		// << std::setprecision(16)
226		// << " calculated: " << calculated << "\n"
227		// << " expected: " << expected << "\n"
228		// << " tolerance: " << tolerance);
229		// calculated = s.skewness();
230		// if (std::fabs(calculated-expected) > tolerance)
231		// BOOST_FAIL("RiskStatistics: wrong skewness"
232		// << " for N(" << averages[i] << ", "
233		// << sigmas[j] << ")\n"
234		// << std::setprecision(16)
235		// << " calculated: " << calculated << "\n"
236		// << " expected: " << expected << "\n"
237		// << " tolerance: " << tolerance);
238
239
240		// // kurtosis
241		// expected = 0.0;
242		// tolerance = 1.0e-1;
243		// calculated = igs.kurtosis();
244		// if (std::fabs(calculated-expected) > tolerance)
245		// BOOST_FAIL("IncrementalGaussianStatistics: "
246		// << "wrong kurtosis"
247		// << " for N(" << averages[i] << ", "
248		// << sigmas[j] << ")\n"
249		// << std::setprecision(16)
250		// << " calculated: " << calculated << "\n"
251		// << " expected: " << expected << "\n"
252		// << " tolerance: " << tolerance);
253		// calculated = s.kurtosis();
254		// if (std::fabs(calculated-expected) > tolerance)
255		// BOOST_FAIL("RiskStatistics: wrong kurtosis"
256		// << " for N(" << averages[i] << ", "
257		// << sigmas[j] << ")\n"
258		// << std::setprecision(16)
259		// << " calculated: " << calculated << "\n"
260		// << " expected: " << expected << "\n"
261		// << " tolerance: " << tolerance);
262
263
264		// // percentile
265		// expected = averages[i];
266		// tolerance = (expected == 0.0 ? 1.0e-3 :
267		// std::fabs(expected*1.0e-3));
268		// calculated = igs.gaussianPercentile(0.5);
269		// if (std::fabs(calculated-expected) > tolerance)
270		// BOOST_FAIL("IncrementalGaussianStatistics: "
271		// << "wrong Gaussian percentile"
272		// << " for N(" << averages[i] << ", "
273		// << sigmas[j] << ")\n"
274		// << std::setprecision(16)
275		// << " calculated: " << calculated << "\n"
276		// << " expected: " << expected << "\n"
277		// << " tolerance: " << tolerance);
278		// calculated = s.gaussianPercentile(0.5);
279		// if (std::fabs(calculated-expected) > tolerance)
280		// BOOST_FAIL("RiskStatistics: wrong Gaussian percentile"
281		// << " for N(" << averages[i] << ", "
282		// << sigmas[j] << ")\n"
283		// << std::setprecision(16)
284		// << " calculated: " << calculated << "\n"
285		// << " expected: " << expected << "\n"
286		// << " tolerance: " << tolerance);
287		// calculated = s.percentile(0.5);
288		// if (std::fabs(calculated-expected) > tolerance)
289		// BOOST_FAIL("RiskStatistics: wrong percentile"
290		// << " for N(" << averages[i] << ", "
291		// << sigmas[j] << ")\n"
292		// << std::setprecision(16)
293		// << " calculated: " << calculated << "\n"
294		// << " expected: " << expected << "\