多変数の極値問題は解析と線形代数の融合だ
•
4 likes•1,873 views
極値を求める問題は工学系に限らずあらゆる面で利用価値のある数学の応用分野である。高校生のときには1変数の場合の極値を取り扱い視覚的にもわかりやすく理解しやすかった。しかし、大学で多変数関数の微分に入ると急に複雑に感じ始める。2変数までは視覚的に捉えられても、3変数以上になるともう感覚的に捉えることができなくなり、理解しにくくなる。極値問題を取り扱うときにも同様で、2変数まではなんとか説明されている大学の教科書も3変数以上になると省略されていることが多い。しかし、実用面から考えると3変数以上の極値を求める応用分野はとても広い。重要なテーマなのである。 ここでは主に3変数以上の極値問題をできるだけわかりやすく概観することを試みる。解析だけの知識だけでは2変数の極値問題に限られてしまうので線形代数の知識も使いながら3変数以上に汎用性を広げていく。線形代数の、行列式、固有値、固有ベクトル、直行変換、2次形式、正値（負値）などの概念を使うことになるが、逆に解析と線形代数のつながりが見えてくる重要な場面にもなっている。
Report
Share
Report
Share
Recommended
コーヒーはホットかアイスか 意外なことが分かった
暑いとアイスコーヒーの注文が多い
寒いとホットコーヒーの注文が多い
こう思うが、果たして正しいのだろうか
そこで、真夏の5日間と真冬の5日間の
ホットコーヒー、アイスコーヒー
の売り上げを調べて季節とこれらの関係を見ることにした
その結果、意外なことが分かった
統計の世界：予測を扱う科学 Statistics World: A Science of Prediction
中学3年生から高校2年生までを対象として統計の世界を紹介する講演の概要である。まず、80段のゴルトンボードを用いて、身の回りで観測される数値には確率的なゆらぎがあることを実験的に確認した。中心極限定理にも触れた。次に、感染症拡大予測についてさまざまな予測手法を紹介し、平均的なふるまいとゆらぎについて説明した。これは、広島大学附属中学校・高等学校での2017年10月27日に行われたSSH特別講義の内容である。
Derivative of sine function: A graphical explanation
The derivative of a sine function can be derived by using the limit for sine function. However, it seems difficult to understand this transformation. Thus, I have drawn a figure expressing the differentiation.
sine関数微分 d sin x / dx = cos x の説明は、sineの差の公式を積に変換して、sin x / x → 1 (x → 0) を使って説明されることが多い。
ここでは、図形的に示してみた。sin x / x → 1 (x → 0) が見えないだけになっているが、結局は、、、
Success/Failure Prediction for Final Examination using the Trend of Weekly On...
H. Hirose, Success/Failure Prediction for Final Examination using the Trend of Weekly Online Testing, 7th International Conference on Learning Technologies and Learning Environments (LTLE2018), pp.139-145, July 8-12, 2018.
Attendance to Lectures is Crucial in Order Not to Drop Out
H. Hirose, Attendance to Lectures is Crucial in Order Not to Drop Out, 7th International Conference on Learning Technologies and Learning Environments (LTLE2018), pp.194-198, July 8-12, 2018.
HTT vs. HTH
How many times are we tossing coins until we observe head, tail, and head? It's ten. It's not eight. This is an intriguing result against our intuition.
Recommended
コーヒーはホットかアイスか 意外なことが分かった
暑いとアイスコーヒーの注文が多い
寒いとホットコーヒーの注文が多い
こう思うが、果たして正しいのだろうか
そこで、真夏の5日間と真冬の5日間の
ホットコーヒー、アイスコーヒー
の売り上げを調べて季節とこれらの関係を見ることにした
その結果、意外なことが分かった
統計の世界：予測を扱う科学 Statistics World: A Science of Prediction
中学3年生から高校2年生までを対象として統計の世界を紹介する講演の概要である。まず、80段のゴルトンボードを用いて、身の回りで観測される数値には確率的なゆらぎがあることを実験的に確認した。中心極限定理にも触れた。次に、感染症拡大予測についてさまざまな予測手法を紹介し、平均的なふるまいとゆらぎについて説明した。これは、広島大学附属中学校・高等学校での2017年10月27日に行われたSSH特別講義の内容である。
Derivative of sine function: A graphical explanation
The derivative of a sine function can be derived by using the limit for sine function. However, it seems difficult to understand this transformation. Thus, I have drawn a figure expressing the differentiation.
sine関数微分 d sin x / dx = cos x の説明は、sineの差の公式を積に変換して、sin x / x → 1 (x → 0) を使って説明されることが多い。
ここでは、図形的に示してみた。sin x / x → 1 (x → 0) が見えないだけになっているが、結局は、、、
Success/Failure Prediction for Final Examination using the Trend of Weekly On...
H. Hirose, Success/Failure Prediction for Final Examination using the Trend of Weekly Online Testing, 7th International Conference on Learning Technologies and Learning Environments (LTLE2018), pp.139-145, July 8-12, 2018.
Attendance to Lectures is Crucial in Order Not to Drop Out
H. Hirose, Attendance to Lectures is Crucial in Order Not to Drop Out, 7th International Conference on Learning Technologies and Learning Environments (LTLE2018), pp.194-198, July 8-12, 2018.
HTT vs. HTH
How many times are we tossing coins until we observe head, tail, and head? It's ten. It's not eight. This is an intriguing result against our intuition.
Homotopy法による非線形方程式の解法
非線形方程式の解法にはNewton法が定番。しかし、収束は初期値に敏感であることはあまり教科書に載っていない。非常に簡単な問題でもいいので試しに指数関数が入った方程式をNewton法で解いてみるとイライラしてくるくらい初期値設定が難しいことがわかる。現実問題とはそんなもの。初期値設定を手助けする意味でHomotopy法を使ってみる。数学的には美しい理論。おもしろいのはHomotopy法を使うということは微分方程式を解いていることにつながるということ。エンジョイして欲しい。
Different classification results under different criteria, distance and proba...
分類の規準を距離の規準にとるか確率の規準にとるかで分類結果が違う。
ここではその典型例示した。
漸近理論をスライド1枚で（フォローアッププログラムクラス講義07132016）
統計学の理論の根幹をなしている「漸近理論」はR.A. Fisherによって打ち立てられた壮大な理論体系です。普通、大学の高学年や大学院で教える内容です。それをここでは大学1年生にも分かるように簡潔に要点のみをスライド1ページで説明することを試みました。
雷の波形は指数関数（フォローアッププログラムクラス講義07072016）
自然界によく観られる現象は指数関数や対数関数で表されることが多いのです。ここでは雷の波形を単純な2つの指数関数の差で表して微分、極値、変曲点を学びましょう。
微分は約分ではない（フォローアッププログラムクラス講義06152016）
2016年前期フォローアッププログラムクラスで行った講義です。微分は約分に似ていますが極限をとりますので意味を考えて使い分けましょう。
Interesting but difficult problem: find the optimum saury layout on a gridiro...
Even though we use a simple but ridiculous problem finding the optimum saury baking layout on a fish gridiron by Joule heat, we can invoke the interest to science by combining electrical engineering, linear algebra and probability viewpoints. These elements are, use of solving linear equation and Poisson's equation, and applying the central limit theorem to this situation. In addition, by removing the constraints, we can create a new problem free from our common sense. Presenting funny but essential problems could be another aspect for active learning using the problem of the interdisciplinary scientific methods.
Central Limit Theorem & Galton Board
With 80 steps Galton boards, we can see the binomial distribution approximated to the normal distribution.
Youtube ===>>>
https://www.youtube.com/watch?v=3w4e1RQTAB8
回路方程式, ポアソン方程式, 中心極限定理による最適なサンマの焼き方
サンマが一尾。魚焼きの網はあるがコンロと炭がない。電気はある。
電気でサンマを焼いてみよう。
回路方程式, ポアソン方程式, 中心極限定理を用いて魚焼き網の両端から電流を流してジュール熱で魚を焼く。
さて、魚をどのように置いたらうまく焼けるだろうか。
最適な焼き方について考えてみよう。
最後にはあっと驚く発想の転換と意外な結果が。
….. 電気学会教育フロンティア研究会, FIE-16-015, pp.71-78（February 29-30, 2016）
Extended cumulative exposure model, ecem
The cumulative exposure model (CEM) is often used to express the failure probability model in the step-up test method; the step-up procedure continues until a breakdown occurs. This probability model is widely accepted in reliability fields because accumulation of fatigue is considered to be reasonable. Contrary to this, the memoryless model (MM) is also used in electrical engineering because accumulation of fatigue is not observed in some cases. We propose here a new model, the extended cumulative exposure model (ECEM), which includes features of both the described models. A simulation study and an application to the actual experimental case of oil insulation test support the validity of the proposed model. The independence model (IM) is also discussed.
Parameter estimation for the truncated weibull model using the ordinary diffe...
In estimating the number of failures using the truncated data for the Weibull model, we often encounter a case that the estimate is smaller than the true one when we use the likelihood principle to conditional probability. In infectious disease predictions, the SIR model described by simultaneous ordinary differential equations are often used, and this model can predict the final stage condition, i.e., the total number of infected patients, well, even if the number of observed data is small. These two models have the same condition for the observed data: truncated to the right. Thus, we have investigated whether the number of failures in the Weibull model can be estimated accurately using the ordinary differential equation. The positive results to this conjecture are shown.
Trunsored data analysis
A new incomplete data model, the trunsored model, in lifetime analysis is introduced. This model can be regarded as a unified model of the censored and truncated models. Using the model, we can not only estimate the ratio of the fragile population to the mixed fragile and durable populations, but also test a hypothesis that the ratio is equal to a prescribed value. A central point of the paper is that such a test can easily be realized through the newly introduced trunsored model, because it has been difficult to do such a hypothesis test under only the framework of censored and truncated models. Therefore, the relationship of the trunsored model to the censored and truncated models is clarified because the trunsored model unifies the censored and truncated models. The paper also shows how to obtain the estimates of the parameters in lifetime estimation, and corresponding confidence intervals for the fragile population. Typical examples applied to electronic board failures, and to breast cancer data, for lifetime estimation are demonstrated, and successfully worked using the trunsored model.
An accurate ability evaluation method for every student with small problem it...
To enhance the chance of use of the item response theory (IRT) in universities, we developed a test evaluation system via the Web for university teachers, and we have been evaluating students' abilities by using the IRT system in midterm and final examinations for two years.
We show a surprising aspect regarding the adoption of the IRT system in university tests. That is, the IRT can not only give us the problem difficulty information but also can provide the accurate student ability evaluation, even if the number of problems is small. Therefore, we can include high and low level test items together so that we can assess a variety of students' abilities accurately and fairly; we do not worry about providing easier problems that will make the lecture level decline; in other words, we do not care about finding the most appropriate problem levels to each student. We can provide all level problems uniformly distributed to all students, and we can still assess the students' abilities accurately. Consequently, students do not raise claims about their scores; they seem to be satisfied with it.
We show these results, in this paper, by a theoretical background, a simulation study, and our empirical results.
A successful maximum likelihood parameter estimation in skewed distributions ...
A successful maximum likelihood parameter estimation scheme using
the continuation method (homotopy method) is introduced. This
algorithm is particularly useful for the three-parameter skewed
distributions including thresholds. Such three-parameter
distributions are, for example, Weibull, log-normal, gamma and
inverse Gaussian distributions. As the proposed algorithm can almost
always obtain the local maximum likelihood estimates automatically,
it is of considerable practical value. The Monte Carlo simulation
study shows the effectiveness of the proposed method.
Estimation for the number of fragile samples in the trunsored and truncated m...
A method to obtain the estimate and its confidence interval for the number of fragile samples in mixed populations of the fragile and durable samples, i.e., in the trunsored model, is introduced. The confidence interval in the trunsored model is compared with that in the truncated model. Although the maximum likelihood estimates for the parameters in the underlying probability distribution in both models are the same, the confidence interval for the estimated number of samples in the trunsored model is differ from that in the truncated model. When the censoring time goes to infinity, the confidence interval in the truncated model converges to zero, whereas the confidence interval in the trunsored model converges to a positive constant value.
The error for the number of fragile samples in the trunsored model is affected by the two kinds of fluctuation effect due to the censoring time: one is the fluctuation of the parameter estimates, and the other is the ratio of the number of fragile samples to the total number of samples. However, in the truncated model, the fluctuation depends only on the parameter estimates, and the error by this effect will vanish when the censoring time goes to infinity.
A typical example of the method is applied to the case fatality ratio for the infectious diseases such as SARS.
Bump hunting とその応用
In difficult classification problems of the z-dimensional points into two groups having 0-1 responses due to the messy data structure, it is more favorable to search for the denser regions for the re- sponse 1 assigned points than to find the boundaries to separate the two groups. To such problems of- ten seen in customer databases, we have developed a bump hunting method using probabilistic and sta- tistical methods. By specifying a pureness rate in advance, a maximum capture rate will be obtained. Then, a trade-off curve between the pureness rate and the capture rate can be constructed. In find- ing the maximum capture rate, we have used the decision tree method combined with the genetic al- gorithm. We first explain a brief introduction of our research: what the bump hunting is, the trade-off curve between the pureness rate and the capture rate, the bump hunting using the tree genetic algorithm, the upper bounds for the trade-off curve using the extreme-value statistics. Then, the assessment for the accuracy of the trade-off curve is tackled from the genetic algorithm procedure viewpoint. Using the new genetic algorithm procedure proposed, we can obtain the upper bound accuracy for the trade- off curve. Then, we may expect the actually attain- able trade-off curve upper bound. The bootstrapped hold-out method is used in assessing the accuracy of the trade-off curve, as well as the cross validation method.
Accuracy assessment for the trade off curve and its upper curve in the bump h...
Suppose that we are interested in classifying n points in a z-dimensional space into two groups having response 1 and response 0 as the target variable. In some real data cases in customer classification, it is difficult to discriminate the favorable customers showing response 1 from others because many re- sponse 1 points and 0 points are closely located. In such a case, to find the denser regions to the favorable customers is considered to be an alter- native. Such regions are called the bumps, and finding them is called the bump hunting. By pre-specifying a pureness rate p in advance a maximum capture rate c could be obtained; the pureness rate is the ratio of the num- ber of response 1 points to the total number of points in the target region; the capture rate is the ratio of the number of response 1 points to the total number of points in the total regions. Then a trade-off curve between p and c can be constructed. Thus, the bump hunting is the same as the trade-off curve constructing. In order to make future actions easier, we adopt simpler boundary shapes for the bumps such as the union of z-dimensional boxes located parallel to some explanation variable axes; this means that we adopt the binary decision tree. Since the conventional binary decision tree will not provide the maximum capture rates because of its local optimizer property, some probabilistic methods would be required. Here, we use the genetic al- gorithm (GA) specified to the tree structure to accomplish this; we call this the tree GA. The tree GA has a tendency to provide many local maxima of the capture rates unlike the ordinary GA. According to this property, we can estimate the upper bound curve for the trade-off curve by using the extreme-value statistics. However, these curves could be optimistic if they are constructed using the training data alone. We should be careful in as- sessing the accuracy of these curves. By applying the test data, the accuracy of the trade-off curve itself can easily be assessed. However, the property of the local maxima would not be preserved. In this paper, we have developed a new tree GA to preserve the property of the local maxima of the capture rates by assessing the test data results in each evolution procedure. Then, the accuracy of the trade-off curve and its upper bound curve are assessed.
Random number generation for the generalized normal distribution using the re...
When we want to grasp the characteristics of the time series signals emitted massively from electric
power apparatuses or electroencephalogram, and want to decide some diagnoses about the apparatuses or
human brains, we may use some statistical distribution functions. In such cases, the generalized normal
distribution is frequently used in pattern analysts. In assessing the correctness of the estimates of the
shape of the distribution function accurately, we often use a Monte Carlo simulation study; thus, a fast
and efficient random number generation method for the distribution function is needed. However, the
method for generating the random numbers of the distribution seems not easy and not yet to have been
developed. In this paper, we propose a random number generation method for the distribution function
using the the rejection method. A newly developed modified adaptive rejection method works well in the case
of log-convex density functions.
More Related Content
More from Hideo Hirose
Homotopy法による非線形方程式の解法
非線形方程式の解法にはNewton法が定番。しかし、収束は初期値に敏感であることはあまり教科書に載っていない。非常に簡単な問題でもいいので試しに指数関数が入った方程式をNewton法で解いてみるとイライラしてくるくらい初期値設定が難しいことがわかる。現実問題とはそんなもの。初期値設定を手助けする意味でHomotopy法を使ってみる。数学的には美しい理論。おもしろいのはHomotopy法を使うということは微分方程式を解いていることにつながるということ。エンジョイして欲しい。
Different classification results under different criteria, distance and proba...
分類の規準を距離の規準にとるか確率の規準にとるかで分類結果が違う。
ここではその典型例示した。
漸近理論をスライド1枚で（フォローアッププログラムクラス講義07132016）
統計学の理論の根幹をなしている「漸近理論」はR.A. Fisherによって打ち立てられた壮大な理論体系です。普通、大学の高学年や大学院で教える内容です。それをここでは大学1年生にも分かるように簡潔に要点のみをスライド1ページで説明することを試みました。
雷の波形は指数関数（フォローアッププログラムクラス講義07072016）
自然界によく観られる現象は指数関数や対数関数で表されることが多いのです。ここでは雷の波形を単純な2つの指数関数の差で表して微分、極値、変曲点を学びましょう。
微分は約分ではない（フォローアッププログラムクラス講義06152016）
2016年前期フォローアッププログラムクラスで行った講義です。微分は約分に似ていますが極限をとりますので意味を考えて使い分けましょう。
Interesting but difficult problem: find the optimum saury layout on a gridiro...
Even though we use a simple but ridiculous problem finding the optimum saury baking layout on a fish gridiron by Joule heat, we can invoke the interest to science by combining electrical engineering, linear algebra and probability viewpoints. These elements are, use of solving linear equation and Poisson's equation, and applying the central limit theorem to this situation. In addition, by removing the constraints, we can create a new problem free from our common sense. Presenting funny but essential problems could be another aspect for active learning using the problem of the interdisciplinary scientific methods.
Central Limit Theorem & Galton Board
With 80 steps Galton boards, we can see the binomial distribution approximated to the normal distribution.
Youtube ===>>>
https://www.youtube.com/watch?v=3w4e1RQTAB8
回路方程式, ポアソン方程式, 中心極限定理による最適なサンマの焼き方
サンマが一尾。魚焼きの網はあるがコンロと炭がない。電気はある。
電気でサンマを焼いてみよう。
回路方程式, ポアソン方程式, 中心極限定理を用いて魚焼き網の両端から電流を流してジュール熱で魚を焼く。
さて、魚をどのように置いたらうまく焼けるだろうか。
最適な焼き方について考えてみよう。
最後にはあっと驚く発想の転換と意外な結果が。
….. 電気学会教育フロンティア研究会, FIE-16-015, pp.71-78（February 29-30, 2016）
Extended cumulative exposure model, ecem
The cumulative exposure model (CEM) is often used to express the failure probability model in the step-up test method; the step-up procedure continues until a breakdown occurs. This probability model is widely accepted in reliability fields because accumulation of fatigue is considered to be reasonable. Contrary to this, the memoryless model (MM) is also used in electrical engineering because accumulation of fatigue is not observed in some cases. We propose here a new model, the extended cumulative exposure model (ECEM), which includes features of both the described models. A simulation study and an application to the actual experimental case of oil insulation test support the validity of the proposed model. The independence model (IM) is also discussed.
Parameter estimation for the truncated weibull model using the ordinary diffe...
In estimating the number of failures using the truncated data for the Weibull model, we often encounter a case that the estimate is smaller than the true one when we use the likelihood principle to conditional probability. In infectious disease predictions, the SIR model described by simultaneous ordinary differential equations are often used, and this model can predict the final stage condition, i.e., the total number of infected patients, well, even if the number of observed data is small. These two models have the same condition for the observed data: truncated to the right. Thus, we have investigated whether the number of failures in the Weibull model can be estimated accurately using the ordinary differential equation. The positive results to this conjecture are shown.
Trunsored data analysis
A new incomplete data model, the trunsored model, in lifetime analysis is introduced. This model can be regarded as a unified model of the censored and truncated models. Using the model, we can not only estimate the ratio of the fragile population to the mixed fragile and durable populations, but also test a hypothesis that the ratio is equal to a prescribed value. A central point of the paper is that such a test can easily be realized through the newly introduced trunsored model, because it has been difficult to do such a hypothesis test under only the framework of censored and truncated models. Therefore, the relationship of the trunsored model to the censored and truncated models is clarified because the trunsored model unifies the censored and truncated models. The paper also shows how to obtain the estimates of the parameters in lifetime estimation, and corresponding confidence intervals for the fragile population. Typical examples applied to electronic board failures, and to breast cancer data, for lifetime estimation are demonstrated, and successfully worked using the trunsored model.
An accurate ability evaluation method for every student with small problem it...
To enhance the chance of use of the item response theory (IRT) in universities, we developed a test evaluation system via the Web for university teachers, and we have been evaluating students' abilities by using the IRT system in midterm and final examinations for two years.
We show a surprising aspect regarding the adoption of the IRT system in university tests. That is, the IRT can not only give us the problem difficulty information but also can provide the accurate student ability evaluation, even if the number of problems is small. Therefore, we can include high and low level test items together so that we can assess a variety of students' abilities accurately and fairly; we do not worry about providing easier problems that will make the lecture level decline; in other words, we do not care about finding the most appropriate problem levels to each student. We can provide all level problems uniformly distributed to all students, and we can still assess the students' abilities accurately. Consequently, students do not raise claims about their scores; they seem to be satisfied with it.
We show these results, in this paper, by a theoretical background, a simulation study, and our empirical results.
A successful maximum likelihood parameter estimation in skewed distributions ...
A successful maximum likelihood parameter estimation scheme using
the continuation method (homotopy method) is introduced. This
algorithm is particularly useful for the three-parameter skewed
distributions including thresholds. Such three-parameter
distributions are, for example, Weibull, log-normal, gamma and
inverse Gaussian distributions. As the proposed algorithm can almost
always obtain the local maximum likelihood estimates automatically,
it is of considerable practical value. The Monte Carlo simulation
study shows the effectiveness of the proposed method.
Estimation for the number of fragile samples in the trunsored and truncated m...
A method to obtain the estimate and its confidence interval for the number of fragile samples in mixed populations of the fragile and durable samples, i.e., in the trunsored model, is introduced. The confidence interval in the trunsored model is compared with that in the truncated model. Although the maximum likelihood estimates for the parameters in the underlying probability distribution in both models are the same, the confidence interval for the estimated number of samples in the trunsored model is differ from that in the truncated model. When the censoring time goes to infinity, the confidence interval in the truncated model converges to zero, whereas the confidence interval in the trunsored model converges to a positive constant value.
The error for the number of fragile samples in the trunsored model is affected by the two kinds of fluctuation effect due to the censoring time: one is the fluctuation of the parameter estimates, and the other is the ratio of the number of fragile samples to the total number of samples. However, in the truncated model, the fluctuation depends only on the parameter estimates, and the error by this effect will vanish when the censoring time goes to infinity.
A typical example of the method is applied to the case fatality ratio for the infectious diseases such as SARS.
Bump hunting とその応用
In difficult classification problems of the z-dimensional points into two groups having 0-1 responses due to the messy data structure, it is more favorable to search for the denser regions for the re- sponse 1 assigned points than to find the boundaries to separate the two groups. To such problems of- ten seen in customer databases, we have developed a bump hunting method using probabilistic and sta- tistical methods. By specifying a pureness rate in advance, a maximum capture rate will be obtained. Then, a trade-off curve between the pureness rate and the capture rate can be constructed. In find- ing the maximum capture rate, we have used the decision tree method combined with the genetic al- gorithm. We first explain a brief introduction of our research: what the bump hunting is, the trade-off curve between the pureness rate and the capture rate, the bump hunting using the tree genetic algorithm, the upper bounds for the trade-off curve using the extreme-value statistics. Then, the assessment for the accuracy of the trade-off curve is tackled from the genetic algorithm procedure viewpoint. Using the new genetic algorithm procedure proposed, we can obtain the upper bound accuracy for the trade- off curve. Then, we may expect the actually attain- able trade-off curve upper bound. The bootstrapped hold-out method is used in assessing the accuracy of the trade-off curve, as well as the cross validation method.
Accuracy assessment for the trade off curve and its upper curve in the bump h...
Suppose that we are interested in classifying n points in a z-dimensional space into two groups having response 1 and response 0 as the target variable. In some real data cases in customer classification, it is difficult to discriminate the favorable customers showing response 1 from others because many re- sponse 1 points and 0 points are closely located. In such a case, to find the denser regions to the favorable customers is considered to be an alter- native. Such regions are called the bumps, and finding them is called the bump hunting. By pre-specifying a pureness rate p in advance a maximum capture rate c could be obtained; the pureness rate is the ratio of the num- ber of response 1 points to the total number of points in the target region; the capture rate is the ratio of the number of response 1 points to the total number of points in the total regions. Then a trade-off curve between p and c can be constructed. Thus, the bump hunting is the same as the trade-off curve constructing. In order to make future actions easier, we adopt simpler boundary shapes for the bumps such as the union of z-dimensional boxes located parallel to some explanation variable axes; this means that we adopt the binary decision tree. Since the conventional binary decision tree will not provide the maximum capture rates because of its local optimizer property, some probabilistic methods would be required. Here, we use the genetic al- gorithm (GA) specified to the tree structure to accomplish this; we call this the tree GA. The tree GA has a tendency to provide many local maxima of the capture rates unlike the ordinary GA. According to this property, we can estimate the upper bound curve for the trade-off curve by using the extreme-value statistics. However, these curves could be optimistic if they are constructed using the training data alone. We should be careful in as- sessing the accuracy of these curves. By applying the test data, the accuracy of the trade-off curve itself can easily be assessed. However, the property of the local maxima would not be preserved. In this paper, we have developed a new tree GA to preserve the property of the local maxima of the capture rates by assessing the test data results in each evolution procedure. Then, the accuracy of the trade-off curve and its upper bound curve are assessed.
Random number generation for the generalized normal distribution using the re...
When we want to grasp the characteristics of the time series signals emitted massively from electric
power apparatuses or electroencephalogram, and want to decide some diagnoses about the apparatuses or
human brains, we may use some statistical distribution functions. In such cases, the generalized normal
distribution is frequently used in pattern analysts. In assessing the correctness of the estimates of the
shape of the distribution function accurately, we often use a Monte Carlo simulation study; thus, a fast
and efficient random number generation method for the distribution function is needed. However, the
method for generating the random numbers of the distribution seems not easy and not yet to have been
developed. In this paper, we propose a random number generation method for the distribution function
using the the rejection method. A newly developed modified adaptive rejection method works well in the case
of log-convex density functions.
More from Hideo Hirose (20)
Different classification results under different criteria, distance and proba...
Different classification results under different criteria, distance and proba...
Interesting but difficult problem: find the optimum saury layout on a gridiro...
Interesting but difficult problem: find the optimum saury layout on a gridiro...
Parameter estimation for the truncated weibull model using the ordinary diffe...
Parameter estimation for the truncated weibull model using the ordinary diffe...
An accurate ability evaluation method for every student with small problem it...
An accurate ability evaluation method for every student with small problem it...
A successful maximum likelihood parameter estimation in skewed distributions ...
A successful maximum likelihood parameter estimation in skewed distributions ...
Estimation for the number of fragile samples in the trunsored and truncated m...
Estimation for the number of fragile samples in the trunsored and truncated m...
Accuracy assessment for the trade off curve and its upper curve in the bump h...
Accuracy assessment for the trade off curve and its upper curve in the bump h...
Random number generation for the generalized normal distribution using the re...
Random number generation for the generalized normal distribution using the re...