{"id":16575,"date":"2019-12-31T19:21:34","date_gmt":"2020-01-01T00:21:34","guid":{"rendered":"https:\/\/bangla.salearningschool.com\/recent-posts\/?p=16575"},"modified":"2020-02-08T09:42:42","modified_gmt":"2020-02-08T14:42:42","slug":"part-4-some-basic-math-stat-concepts-for-the-wanna-be-data-scientists","status":"publish","type":"post","link":"http:\/\/bangla.sitestree.com\/?p=16575","title":{"rendered":"Part 4: Some Basic Math\/Stat Concepts for the wanna be Data Scientists"},"content":{"rendered":"<p>Part 4: Some Basic Math\/Stat Concepts for the wanna be Data Scientists<\/p>\n<p>Also for the Engineers in General<\/p>\n<h1>Quadratic form<\/h1>\n<p>&#8220;In <a title=\"Multivariate statistics\" href=\"https:\/\/en.wikipedia.org\/wiki\/Multivariate_statistics\">multivariate statistics<\/a>, if <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/a30c89172e5b88edbd45d3e2772c7f5e562e5173\" alt=\"\\varepsilon\" \/> is a <a title=\"Vector space\" href=\"https:\/\/en.wikipedia.org\/wiki\/Vector_space\">vector<\/a> of <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/a601995d55609f2d9f5e233e36fbe9ea26011b3b\" alt=\"n\" \/> <a title=\"Random variable\" href=\"https:\/\/en.wikipedia.org\/wiki\/Random_variable\">random variables<\/a>, and <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/0ac0a4a98a414e3480335f9ba652d12571ec6733\" alt=\"\\Lambda\" \/> is an <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/a601995d55609f2d9f5e233e36fbe9ea26011b3b\" alt=\"n\" \/>-dimensional <a title=\"Symmetric matrix\" href=\"https:\/\/en.wikipedia.org\/wiki\/Symmetric_matrix\">symmetric matrix<\/a>, then the <a title=\"Scalar (mathematics)\" href=\"https:\/\/en.wikipedia.org\/wiki\/Scalar_(mathematics)\">scalar<\/a> quantity <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/c07b9d34b500c61f997da04c838514b081bbf66f\" alt=\"{\\displaystyle \\varepsilon ^{T}\\Lambda \\varepsilon }\" \/> is known as a <strong>quadratic form<\/strong> in <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/a30c89172e5b88edbd45d3e2772c7f5e562e5173\" alt=\"\\varepsilon\" \/>.<br \/>\n&#8221;<\/p>\n<p>Ref: <a href=\"https:\/\/en.wikipedia.org\/wiki\/Quadratic_form_(statistics)\">https:\/\/en.wikipedia.org\/wiki\/Quadratic_form_(statistics)<\/a><\/p>\n<p>Please also check matrix related concepts. We will provide some matrix concepts at one point.<\/p>\n<p>&#8220;In mathematics, a quadratic form is a polynomial with terms all of degree two. For example, is a quadratic form in the variables x and y. <a href=\"https:\/\/en.wikipedia.org\/wiki\/Quadratic_form\">Wikipedia<\/a>&#8221;<\/p>\n<p>&#8221;<br \/>\n<img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/50c3a2e36e367855e6a2107fb18c9c3f75812aaf\" alt=\"4x^2 + 2xy - 3y^2\" \/>is a quadratic form in the variables <em>x<\/em> and <em>y<\/em>. The coefficients usually belong to a fixed field <em>K<\/em>, such as the real or complex numbers, and we speak of a quadratic form over <em>K<\/em>.&#8221;<\/p>\n<p>&#8220;Quadratic forms are not to be confused with a <a title=\"Quadratic equation\" href=\"https:\/\/en.wikipedia.org\/wiki\/Quadratic_equation\">quadratic equation<\/a> which has only one variable and includes terms of degree two or less. A quadratic form is one case of the more general concept of <a title=\"Homogeneous polynomial\" href=\"https:\/\/en.wikipedia.org\/wiki\/Homogeneous_polynomial\">homogeneous polynomials<\/a>.&#8221;<\/p>\n<p>Ref: <a href=\"https:\/\/en.wikipedia.org\/wiki\/Quadratic_form\">https:\/\/en.wikipedia.org\/wiki\/Quadratic_form<\/a><\/p>\n<h1>Quartic function<\/h1>\n<p>&#8221;<br \/>\nThis article is about the univariate case. For the bivariate case, see <a title=\"Quartic plane curve\" href=\"https:\/\/en.wikipedia.org\/wiki\/Quartic_plane_curve\">Quartic plane curve<\/a>.<br \/>\n<a href=\"https:\/\/en.wikipedia.org\/wiki\/File:Polynomialdeg4.svg\"><img loading=\"lazy\" decoding=\"async\" src=\"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/3\/3c\/Polynomialdeg4.svg\/233px-Polynomialdeg4.svg.png\" alt=\"\" width=\"233\" height=\"233\" \/><\/a><br \/>\nGraph of a polynomial of degree 4, with 3 <a title=\"Critical point (mathematics)\" href=\"https:\/\/en.wikipedia.org\/wiki\/Critical_point_(mathematics)\">critical points<\/a> and four <a title=\"Real number\" href=\"https:\/\/en.wikipedia.org\/wiki\/Real_number\">real<\/a> <a title=\"Root of a polynomial\" href=\"https:\/\/en.wikipedia.org\/wiki\/Root_of_a_polynomial\">roots<\/a> (crossings of the <em>x<\/em> axis) (and thus no <a title=\"Complex number\" href=\"https:\/\/en.wikipedia.org\/wiki\/Complex_number\">complex<\/a> roots). If one or the other of the local <a title=\"Minimum\" href=\"https:\/\/en.wikipedia.org\/wiki\/Minimum\">minima<\/a> were above the <em>x<\/em> axis, or if the local <a title=\"Maximum\" href=\"https:\/\/en.wikipedia.org\/wiki\/Maximum\">maximum<\/a> were below it, or if there were no local maximum and one minimum below the <em>x<\/em> axis, there would only be two real roots (and two complex roots). If all three local extrema were above the <em>x<\/em> axis, or if there were no local maximum and one minimum above the <em>x<\/em> axis, there would be no real root (and four complex roots). The same reasoning applies in reverse to polynomial with a negative quartic coefficient.<\/p>\n<p>In <a title=\"Algebra\" href=\"https:\/\/en.wikipedia.org\/wiki\/Algebra\">algebra<\/a>, a <strong>quartic function<\/strong> is a <a title=\"Function (mathematics)\" href=\"https:\/\/en.wikipedia.org\/wiki\/Function_(mathematics)\">function<\/a> of the form<\/p>\n<p><img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/fd9ab03f4edd6e4bf1add76e4c8b27f040ad9bd9\" alt=\"f(x)=ax^{4}+bx^{3}+cx^{2}+dx+e,\" \/><\/p>\n<p>where <em>a<\/em> is nonzero, which is defined by a <a title=\"Polynomial\" href=\"https:\/\/en.wikipedia.org\/wiki\/Polynomial\">polynomial<\/a> of <a title=\"Degree of a polynomial\" href=\"https:\/\/en.wikipedia.org\/wiki\/Degree_of_a_polynomial\">degree<\/a> four, called a <strong>quartic polynomial<\/strong>.<\/p>\n<p>Sometimes the term <strong>biquadratic<\/strong> is used instead of <em>quartic<\/em>, but, usually, <strong>biquadratic function<\/strong> refers to a <a title=\"Quadratic function\" href=\"https:\/\/en.wikipedia.org\/wiki\/Quadratic_function\">quadratic function<\/a> of a square (or, equivalently, to the function defined by a quartic polynomial without terms of odd degree), having the form<\/p>\n<p><img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/2831f8e2df7bd84e71d53bba0f27dc9efbc3e0d5\" alt=\"f(x)=ax^{4}+cx^{2}+e.\" \/><\/p>\n<p>A <strong>quartic equation<\/strong>, or equation of the fourth degree, is an equation that equates a quartic polynomial to zero, of the form<\/p>\n<p><img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/7631042a8e33a24f37db475097852067c6472afe\" alt=\"ax^{4}+bx^{3}+cx^{2}+dx+e=0,\" \/><\/p>\n<p>where <em>a<\/em> \u2260 0.<\/p>\n<p>The <a title=\"Derivative\" href=\"https:\/\/en.wikipedia.org\/wiki\/Derivative\">derivative<\/a> of a quartic function is a <a title=\"Cubic function\" href=\"https:\/\/en.wikipedia.org\/wiki\/Cubic_function\">cubic function<\/a>.<\/p>\n<p>&#8221;<\/p>\n<p>Ref: <a href=\"https:\/\/en.wikipedia.org\/wiki\/Quartic_function\">https:\/\/en.wikipedia.org\/wiki\/Quartic_function<\/a><\/p>\n<h1>Quartic plane curve<\/h1>\n<p>Bivariate case<\/p>\n<p>&#8221;<\/p>\n<p>A <strong>quartic plane curve<\/strong> is a <a title=\"Plane algebraic curve\" href=\"https:\/\/en.wikipedia.org\/wiki\/Plane_algebraic_curve\">plane algebraic curve<\/a> of the fourth <a title=\"Degree of a polynomial\" href=\"https:\/\/en.wikipedia.org\/wiki\/Degree_of_a_polynomial\">degree<\/a>. It can be defined by a bivariate quartic equation:<\/p>\n<p><img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/b89630a2b623a42b13b85c024ef21a2d7c4433c8\" alt=\"Ax^4+By^4+Cx^3y+Dx^2y^2+Exy^3+Fx^3+Gy^3+Hx^2y+Ixy^2+Jx^2+Ky^2+Lxy+Mx+Ny+P=0,\" \/><\/p>\n<p>with at least one of <em>A, B, C, D, E<\/em> not equal to zero. This equation has 15 constants. However, it can be multiplied by any non-zero constant without changing the curve; thus by the choice of an appropriate constant of multiplication, any one of the coefficients can be set to 1, leaving only 14 constants. Therefore, the space of quartic curves can be identified with the <a title=\"Real projective space\" href=\"https:\/\/en.wikipedia.org\/wiki\/Real_projective_space\">real projective space<\/a> <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/1e0105932f1662afeac50384c081e6dec7f48e3b\" alt=\"{\\mathbb {RP}}^{{14}}\" \/>. It also follows, from <a title=\"Cramer's theorem (algebraic curves)\" href=\"https:\/\/en.wikipedia.org\/wiki\/Cramer%27s_theorem_(algebraic_curves)\">Cramer&#8217;s theorem on algebraic curves<\/a>, that there is exactly one quartic curve that passes through a set of 14 distinct points in <a title=\"General position\" href=\"https:\/\/en.wikipedia.org\/wiki\/General_position\">general position<\/a>, since a quartic has 14 <a title=\"Degrees of freedom (physics and chemistry)\" href=\"https:\/\/en.wikipedia.org\/wiki\/Degrees_of_freedom_(physics_and_chemistry)\">degrees of freedom<\/a>.<\/p>\n<p>A quartic curve can have a maximum of:<\/p>\n<ul>\n<li>Four connected components<\/li>\n<li>Twenty-eight <a title=\"Bitangent\" href=\"https:\/\/en.wikipedia.org\/wiki\/Bitangent\">bi-tangents<\/a><\/li>\n<li>Three ordinary <a title=\"Double point\" href=\"https:\/\/en.wikipedia.org\/wiki\/Double_point\">double points<\/a>.<\/li>\n<\/ul>\n<p>&#8221;<\/p>\n<p><strong>Ref<\/strong>: <a href=\"https:\/\/en.wikipedia.org\/wiki\/Quartic_plane_curve\">https:\/\/en.wikipedia.org\/wiki\/Quartic_plane_curve<\/a><\/p>\n<p><strong>Expected value of Quadratic Forms<\/strong><\/p>\n<p>&#8220;In <a title=\"Multivariate statistics\" href=\"https:\/\/en.wikipedia.org\/wiki\/Multivariate_statistics\">multivariate statistics<\/a>, if <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/a30c89172e5b88edbd45d3e2772c7f5e562e5173\" alt=\"\\varepsilon\" \/> is a <a title=\"Vector space\" href=\"https:\/\/en.wikipedia.org\/wiki\/Vector_space\">vector<\/a> of <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/a601995d55609f2d9f5e233e36fbe9ea26011b3b\" alt=\"n\" \/> <a title=\"Random variable\" href=\"https:\/\/en.wikipedia.org\/wiki\/Random_variable\">random variables<\/a>, and <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/0ac0a4a98a414e3480335f9ba652d12571ec6733\" alt=\"\\Lambda\" \/> is an <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/a601995d55609f2d9f5e233e36fbe9ea26011b3b\" alt=\"n\" \/>-dimensional <a title=\"Symmetric matrix\" href=\"https:\/\/en.wikipedia.org\/wiki\/Symmetric_matrix\">symmetric matrix<\/a>, then the <a title=\"Scalar (mathematics)\" href=\"https:\/\/en.wikipedia.org\/wiki\/Scalar_(mathematics)\">scalar<\/a> quantity <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/c07b9d34b500c61f997da04c838514b081bbf66f\" alt=\"{\\displaystyle \\varepsilon ^{T}\\Lambda \\varepsilon }\" \/> is known as a <strong>quadratic form<\/strong> in <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/a30c89172e5b88edbd45d3e2772c7f5e562e5173\" alt=\"\\varepsilon\" \/>.<br \/>\n&#8221;<\/p>\n<p><strong>Expected Value :<\/strong><\/p>\n<p>&#8220;It can be shown that<a href=\"https:\/\/en.wikipedia.org\/wiki\/Quadratic_form_(statistics)#cite_note-1\">[1]<\/a><\/p>\n<p><img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/baad183f5bdae8ceea0ab20ebb804d7767187c36\" alt=\"{\\displaystyle \\operatorname {E} \\left[\\varepsilon ^{T}\\Lambda \\varepsilon \\right]=\\operatorname {tr} \\left[\\Lambda \\Sigma \\right]+\\mu ^{T}\\Lambda \\mu }\" \/><\/p>\n<p>where <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/9fd47b2a39f7a7856952afec1f1db72c67af6161\" alt=\"\\mu\" \/> and <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/9e1f558f53cda207614abdf90162266c70bc5c1e\" alt=\"\\Sigma\" \/> are the <a title=\"Expected value\" href=\"https:\/\/en.wikipedia.org\/wiki\/Expected_value\">expected value<\/a> and <a title=\"Covariance matrix\" href=\"https:\/\/en.wikipedia.org\/wiki\/Covariance_matrix\">variance-covariance matrix<\/a> of <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/a30c89172e5b88edbd45d3e2772c7f5e562e5173\" alt=\"\\varepsilon\" \/>, respectively, and tr denotes the <a title=\"Trace (linear algebra)\" href=\"https:\/\/en.wikipedia.org\/wiki\/Trace_(linear_algebra)\">trace<\/a> of a matrix. This result only depends on the existence of <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/9fd47b2a39f7a7856952afec1f1db72c67af6161\" alt=\"\\mu\" \/> and <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/9e1f558f53cda207614abdf90162266c70bc5c1e\" alt=\"\\Sigma\" \/>; in particular, <a title=\"Multivariate normal distribution\" href=\"https:\/\/en.wikipedia.org\/wiki\/Multivariate_normal_distribution\">normality<\/a> of <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/a30c89172e5b88edbd45d3e2772c7f5e562e5173\" alt=\"\\varepsilon\" \/> is <em>not<\/em> required.<\/p>\n<p>&#8221;<\/p>\n<p>Note: you might see <img decoding=\"async\" src=\"https:\/\/wikimedia.org\/api\/rest_v1\/media\/math\/render\/svg\/a30c89172e5b88edbd45d3e2772c7f5e562e5173\" alt=\"\\varepsilon\" \/> is replaced with x, and x&#8217; is used for transpose(x).<\/p>\n<p>Also,<\/p>\n<p>may be the equation without the second part (sure there will be an explanation)<\/p>\n<p>The equations above hold irrespective of the distribution of x.<\/p>\n<p><strong>Expected value of Quartic form:<\/strong><br \/>\n<a href=\"https:\/\/i0.wp.com\/bangla.salearningschool.com\/wp-content\/uploads\/2019\/12\/image-18.png\" rel=\"attachment wp-att-16577\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" class=\"alignnone  wp-image-16577\" title=\"image-18-png\" src=\"https:\/\/i0.wp.com\/bangla.salearningschool.com\/wp-content\/uploads\/2019\/12\/image-18.png?resize=576%2C370\" alt=\"\" width=\"576\" height=\"370\" srcset=\"https:\/\/i0.wp.com\/bangla.sitestree.com\/wp-content\/uploads\/2019\/12\/image-18.png?w=617 617w, https:\/\/i0.wp.com\/bangla.sitestree.com\/wp-content\/uploads\/2019\/12\/image-18.png?resize=300%2C193 300w\" sizes=\"auto, (max-width: 576px) 100vw, 576px\" \/><\/a><\/p>\n<p>Ref: Estimation Books by Yaakov Bar-Shalom, X. Rong Li, Thiagalingam Kirubarajan<\/p>\n<p><strong>Mixture Density<\/strong><\/p>\n<p>&#8220;<strong>Mixture distribution<\/strong>. &#8230; In cases where each of the underlying random variables is continuous, the outcome variable will also be continuous and its <strong>probability density function<\/strong> is sometimes referred to as a <strong>mixture density<\/strong>.&#8221;<\/p>\n<p>Ref: <a href=\"https:\/\/en.wikipedia.org\/wiki\/Mixture_distribution\">https:\/\/en.wikipedia.org\/wiki\/Mixture_distribution<\/a><\/p>\n<p><strong>Mixture PDF:<\/strong><\/p>\n<p><strong>&#8220;<\/strong>A mixture pdf is a weighted sum of pdfs with the weights summing up to unity<strong>&#8220;<\/strong><\/p>\n<p>gaussian mixture pdf consists of weighted sum of gaussian densities<\/p>\n<p><a href=\"https:\/\/i0.wp.com\/bangla.salearningschool.com\/wp-content\/uploads\/2019\/12\/image-19.png\" rel=\"attachment wp-att-16578\"><img data-recalc-dims=\"1\" loading=\"lazy\" decoding=\"async\" class=\" wp-image-16578 aligncenter\" title=\"image-19-png\" src=\"https:\/\/i0.wp.com\/bangla.salearningschool.com\/wp-content\/uploads\/2019\/12\/image-19.png?resize=610%2C472\" alt=\"\" width=\"610\" height=\"472\" srcset=\"https:\/\/i0.wp.com\/bangla.sitestree.com\/wp-content\/uploads\/2019\/12\/image-19.png?w=638 638w, https:\/\/i0.wp.com\/bangla.sitestree.com\/wp-content\/uploads\/2019\/12\/image-19.png?resize=300%2C232 300w\" sizes=\"auto, (max-width: 610px) 100vw, 610px\" \/><\/a>Ref: <a href=\"https:\/\/www.slideshare.net\/jins0618\/clusteringkmeans-expectmaximization-and-gaussian-mixture-model\">https:\/\/www.slideshare.net\/jins0618\/clusteringkmeans-expectmaximization-and-gaussian-mixture-model<\/a><\/p>\n<p><a href=\"https:\/\/www.mathworks.com\/help\/stats\/gmdistribution.pdf.html\">https:\/\/www.mathworks.com\/help\/stats\/gmdistribution.pdf.html<\/a><\/p>\n<p><a href=\"http:\/\/digitalcommons.utep.edu\/cgi\/viewcontent.cgi?article=2110&amp;context=cs_techrep\">http:\/\/digitalcommons.utep.edu\/cgi\/viewcontent.cgi?article=2110&amp;context=cs_techrep<\/a><\/p>\n<p>ML and Mixture Models:<\/p>\n<p><a href=\"https:\/\/www.cs.toronto.edu\/~rgrosse\/csc321\/mixture_models.pdf\">https:\/\/www.cs.toronto.edu\/~rgrosse\/csc321\/mixture_models.pdf<\/a><\/p>\n<p><a href=\"https:\/\/statweb.stanford.edu\/~tibs\/stat315a\/LECTURES\/em.pdf\">https:\/\/statweb.stanford.edu\/~tibs\/stat315a\/LECTURES\/em.pdf<\/a><\/p>\n<p>Definitions: <a href=\"https:\/\/www.statisticshowto.datasciencecentral.com\/mixture-distribution\/\">https:\/\/www.statisticshowto.datasciencecentral.com\/mixture-distribution\/<\/a><\/p>\n<p><a href=\"https:\/\/www.asc.ohio-state.edu\/gan.1\/teaching\/spring04\/Chapter3.pdf\">https:\/\/www.asc.ohio-state.edu\/gan.1\/teaching\/spring04\/Chapter3.pdf<\/a><\/p>\n<p>************<br \/>\n<em><strong>Sayed Ahmed<\/strong><br \/>\n<\/em><br \/>\n<em><strong>BSc. Eng. in Comp. Sc. &amp; Eng. (BUET)<\/strong><\/em><br \/>\n<em><strong>MSc. in Comp. Sc. (U of Manitoba, Canada)<\/strong><\/em><br \/>\n<em><strong>MSc. in Data Science and Analytics (Ryerson University, Canada)<\/strong><\/em><br \/>\n<em><strong>Linkedin<\/strong>: <a href=\"https:\/\/ca.linkedin.com\/in\/sayedjustetc\">https:\/\/ca.linkedin.com\/in\/sayedjustetc<\/a><br \/>\n<\/em><\/p>\n<p><em><strong>Blog<\/strong>: <a href=\"http:\/\/bangla.salearningschool.com\/\">http:\/\/Bangla.SaLearningSchool.com<\/a>, <a href=\"http:\/\/sitestree.com\">http:\/\/SitesTree.com<\/a><\/em><br \/>\n<em><strong>Online and Offline Training<\/strong>: <a href=\"http:\/\/training.SitesTree.com\">http:\/\/Training.SitesTree.com<\/a> <\/em><br \/>\n<em>FB Group on Learning\/Teaching: <\/em><a href=\"https:\/\/www.facebook.com\/banglasalearningschool\">https:\/\/www.facebook.com\/banglasalearningschool<\/a><br \/>\nOur free or paid events on IT\/Data Science\/Cloud\/Programming\/Similar: <a href=\"https:\/\/www.facebook.com\/justetcsocial\">https:\/\/www.facebook.com\/justetcsocial<\/a><\/p>\n<p><em>Get access to courses on Big Data, Data Science, AI, Cloud, Linux, System Admin, Web Development and Misc. related. Also, create your own course to sell to others. <\/em><a href=\"http:\/\/sitestree.com\/training\/\">http:\/\/sitestree.com\/training\/<\/a><\/p>\n<p><em><strong>I<\/strong>f you want to contribute to the operation of this site (Bangla.SaLearn) including occasional free and\/or low cost online\/offline training: <a href=\"http:\/\/training.sitestree.com\/\">http:\/\/Training.SitesTree.com<\/a> (or charitable\/non-profit work in the education\/health\/social service sector), you can financially contribute to: safoundation at <a href=\"http:\/\/salearningschool.com\">salearningschool.com<\/a> using Paypal or Credit Card (on <\/em><a href=\"http:\/\/sitestree.com\/training\/enrol\/index.php?id=114\">http:\/\/sitestree.com\/training\/enrol\/index.php?id=114<\/a> <em>).<\/em><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Part 4: Some Basic Math\/Stat Concepts for the wanna be Data Scientists Also for the Engineers in General Quadratic form &#8220;In multivariate statistics, if is a vector of random variables, and is an -dimensional symmetric matrix, then the scalar quantity is known as a quadratic form in . &#8221; Ref: https:\/\/en.wikipedia.org\/wiki\/Quadratic_form_(statistics) Please also check matrix &hellip; <\/p>\n<p><a class=\"more-link btn\" href=\"http:\/\/bangla.sitestree.com\/?p=16575\">Continue reading<\/a><\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_jetpack_memberships_contains_paid_content":false,"footnotes":""},"categories":[1910,1908,182],"tags":[],"class_list":["post-16575","post","type-post","status-publish","format-standard","hentry","category-ai-ml-ds-rl-dl-nn-nlp-data-mining-optimization","category-math-and-statistics-for-data-science-and-engineering","category---blog","item-wrap"],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"jetpack-related-posts":[{"id":16550,"url":"http:\/\/bangla.sitestree.com\/?p=16550","url_meta":{"origin":16575,"position":0},"title":"Part 3: Some Basic Math\/Stat Concepts for the wanna be Data Scientists","author":"Sayed","date":"December 30, 2019","format":false,"excerpt":"Conditional Probability and PDF \"The conditional probability of an event B is the probability that the event will occur given the knowledge that an event A has already occurred. This probability is written P(B|A), notation for the probability of B given A. \" \"In the case where events A and\u2026","rel":"","context":"In &quot;AI ML DS RL DL NN NLP Data Mining Optimization&quot;","block_context":{"text":"AI ML DS RL DL NN NLP Data Mining Optimization","link":"http:\/\/bangla.sitestree.com\/?cat=1910"},"img":{"alt_text":"","src":"https:\/\/i0.wp.com\/bangla.sitestree.com\/wp-content\/uploads\/2019\/12\/image-8.png?resize=350%2C200","width":350,"height":200,"srcset":"https:\/\/i0.wp.com\/bangla.sitestree.com\/wp-content\/uploads\/2019\/12\/image-8.png?resize=350%2C200 1x, https:\/\/i0.wp.com\/bangla.sitestree.com\/wp-content\/uploads\/2019\/12\/image-8.png?resize=525%2C300 1.5x"},"classes":[]},{"id":16536,"url":"http:\/\/bangla.sitestree.com\/?p=16536","url_meta":{"origin":16575,"position":1},"title":"Part 2: Some basic Math\/Statistics concepts that Data Scientists (the true ones) will usually know\/use","author":"Sayed","date":"December 29, 2019","format":false,"excerpt":"Part 2: Some basic Math\/Statistics concepts that Data Scientists (the true ones) will usually know\/use (came across, studied, learned, used) Covariance and Correlation \"Covariance is a measure of how two variables change together, but its magnitude is unbounded, so it is difficult to interpret. By dividing covariance by the product\u2026","rel":"","context":"In &quot;Math and Statistics for Data Science, and Engineering&quot;","block_context":{"text":"Math and Statistics for Data Science, and Engineering","link":"http:\/\/bangla.sitestree.com\/?cat=1908"},"img":{"alt_text":"[eq5]","src":"https:\/\/i0.wp.com\/www.statlect.com\/images\/covariance-formula__12.png?resize=350%2C200&ssl=1","width":350,"height":200},"classes":[]},{"id":16532,"url":"http:\/\/bangla.sitestree.com\/?p=16532","url_meta":{"origin":16575,"position":2},"title":"Part 1: Some Math\/Stat Background that (true) Data Scientists will know\/use: from the internet","author":"Sayed","date":"December 28, 2019","format":false,"excerpt":"Chebyshev's inequality \"In probability theory, Chebyshev's inequality (also called the Bienaym\u00e9\u2013Chebyshev inequality) guarantees that, for a wide class of probability distributions, no more than a certain fraction of values can be more than a certain distance from the mean. Specifically, no more than 1\/k2 of the distribution's values can be\u2026","rel":"","context":"In &quot;Math and Statistics for Data Science, and Engineering&quot;","block_context":{"text":"Math and Statistics for Data Science, and Engineering","link":"http:\/\/bangla.sitestree.com\/?cat=1908"},"img":{"alt_text":"","src":"https:\/\/upload.wikimedia.org\/wikipedia\/commons\/thumb\/8\/85\/Discrete_probability_distrib.svg\/220px-Discrete_probability_distrib.svg.png","width":350,"height":200},"classes":[]},{"id":16619,"url":"http:\/\/bangla.sitestree.com\/?p=16619","url_meta":{"origin":16575,"position":3},"title":"Math\/Stat\/CS\/DS Topics that you need to know (with Cognitive, Psychomotor, Affective domain skills) to become a true and great Data Scientist","author":"Sayed","date":"January 7, 2020","format":false,"excerpt":"\"The core topics are cross-validation, shrinkage methods (ridge regression, the LASSO, etc.), neural networks, gradient boosting, separating hyperplanes, support vector machines, basis expansion and regularization (e.g., smoothing splines, wavelet smoothing, kernel smoothing), generalized additive models, bump hunting, multivariate adaptive regression splines (MARS), self-organizing maps, mixture model-based clustering, ensemble learning, and\u2026","rel":"","context":"In &quot;Math and Statistics for Data Science, and Engineering&quot;","block_context":{"text":"Math and Statistics for Data Science, and Engineering","link":"http:\/\/bangla.sitestree.com\/?cat=1908"},"img":{"alt_text":"","src":"","width":0,"height":0},"classes":[]},{"id":16200,"url":"http:\/\/bangla.sitestree.com\/?p=16200","url_meta":{"origin":16575,"position":4},"title":"Important Basic Concepts: Statistics for Big Data","author":"Sayed","date":"September 15, 2019","format":false,"excerpt":"Important Basic Concepts: Statistics for Big Data Graphical : Exploratory Data Analysis (EDA) methods? First of all, EDA is about exploring the data and understanding if the data will be good for the experiment and study. Graphs and plots can easily show the data patterns. The raw data can be\u2026","rel":"","context":"In &quot;Statistics for Big Data&quot;","block_context":{"text":"Statistics for Big Data","link":"http:\/\/bangla.sitestree.com\/?cat=1904"},"img":{"alt_text":"","src":"","width":0,"height":0},"classes":[]},{"id":26205,"url":"http:\/\/bangla.sitestree.com\/?p=26205","url_meta":{"origin":16575,"position":5},"title":"Important Basic Concepts: Statistics for Big Data #Root","author":"Author-Check- Article-or-Video","date":"April 19, 2021","format":false,"excerpt":"Important Basic Concepts: Statistics for Big Data Graphical : Exploratory Data Analysis (EDA) methods? First of all, EDA is about exploring the data and understanding if the data will be good for the experiment and study. Graphs and plots can easily show the data patterns. The raw data can be\u2026","rel":"","context":"In &quot;FromSitesTree.com&quot;","block_context":{"text":"FromSitesTree.com","link":"http:\/\/bangla.sitestree.com\/?cat=1917"},"img":{"alt_text":"","src":"","width":0,"height":0},"classes":[]}],"_links":{"self":[{"href":"http:\/\/bangla.sitestree.com\/index.php?rest_route=\/wp\/v2\/posts\/16575","targetHints":{"allow":["GET"]}}],"collection":[{"href":"http:\/\/bangla.sitestree.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"http:\/\/bangla.sitestree.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"http:\/\/bangla.sitestree.com\/index.php?rest_route=\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"http:\/\/bangla.sitestree.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=16575"}],"version-history":[{"count":3,"href":"http:\/\/bangla.sitestree.com\/index.php?rest_route=\/wp\/v2\/posts\/16575\/revisions"}],"predecessor-version":[{"id":16581,"href":"http:\/\/bangla.sitestree.com\/index.php?rest_route=\/wp\/v2\/posts\/16575\/revisions\/16581"}],"wp:attachment":[{"href":"http:\/\/bangla.sitestree.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=16575"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"http:\/\/bangla.sitestree.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=16575"},{"taxonomy":"post_tag","embeddable":true,"href":"http:\/\/bangla.sitestree.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=16575"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}