View source for Birth-and-death process

A Markov process with states <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b0165501.png" /> in which in a time interval <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b0165502.png" /> transitions from the state <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b0165503.png" /> into the states <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b0165504.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b0165505.png" /> occur with probabilities <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b0165506.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b0165507.png" />, respectively, and where the probability of other transitions is <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b0165508.png" />. For a special choice of the reproduction coefficients <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b0165509.png" /> and the death coefficients <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655010.png" /> one can obtain particular cases which provide a satisfactory description of various real processes: radioactive transformations, the running of telephone exchanges, the evolution of biological populations, etc. The fact that birth-and-death processes are widely used in applications is due to the simplicity of the equations for the transition probabilities, which often can be obtained explicitly. For instance, in the case of a Poisson process <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655011.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655012.png" />, the probabilities <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655013.png" /> (<img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655014.png" /> is the probability of being at time <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655015.png" /> in the state <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655016.png" /> if the process has started from state <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655017.png" />) satisfy the system:

<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655018.png" /></td> </tr></table>

<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655019.png" /></td> </tr></table>

where

<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655020.png" /></td> </tr></table>

The solution of this system is:

<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655021.png" /></td> </tr></table>

A more general process is the one in which <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655022.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655023.png" />. This type of process was first studied by G. Yule (1924) in connection with the mathematical theory of evolution. A Yule process is a particular case of a pure birth process which is obtained from the general birth-and-death process by assuming <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655024.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655025.png" />. If <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655026.png" /> increases very rapidly, then with positive probability one can pass through all states in a finite time and then

<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655027.png" /></td> </tr></table>

The condition

<table class="eq" style="width:100%;"> <tr><td valign="top" style="width:94%;text-align:center;"><img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655028.png" /></td> </tr></table>

for a pure birth process is satisfied if and only if the series <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655029.png" /> diverges. If <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655030.png" />, <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655031.png" />, then the birth-and-death process is a [[Branching process|branching process]] with immigration in which the state <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655032.png" /> designates the number of particles. Each particle dies during time <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655033.png" /> with probability <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655034.png" />, splits into two particles with probability <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655035.png" /> and, moreover, one particle immigrates from the outside with probability <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655036.png" />. If <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655037.png" /> then one obtains the simplest branching process without immigration. If <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655038.png" /> and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655039.png" />, then this type of process with immigration can be applied to describe the functioning of a telephone exchange with an infinite number of lines. Here, the state is the number of occupied lines. The reproduction coefficient <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655040.png" /> characterizes the rate of incoming telephone calls, and <img align="absmiddle" border="0" src="https://www.encyclopediaofmath.org/legacyimages/b/b016/b016550/b01655041.png" /> is the expected duration of a conversation.

====References====
<table><TR><TD valign="top">[1]</TD> <TD valign="top">  W. Feller, [[Feller, "An introduction to probability theory and its  applications"|"An introduction to probability theory and its  applications"]], '''1–2''' , Wiley  (1957–1971)</TD></TR><TR><TD valign="top">[2]</TD> <TD valign="top">  B.A. [B.A. Sevast'yanov] Sewastjanow,   "Verzweigungsprozesse" , Akad. Wissenschaft. DDR  (1974)  (Translated from Russian)</TD></TR><TR><TD valign="top">[3]</TD> <TD valign="top">  R.L. Saaty,   "Elements of queueing theory with applications" , McGraw-Hill  (1961)</TD></TR></table>

====Comments====


====References====
<table><TR><TD valign="top">[a1]</TD> <TD valign="top">  K.B. Athreya,   P.E. Ney,   "Branching processes" , Springer  (1972)</TD></TR></table>