Basic Examples (1) 

Generate a sequence of 100 options with a bias of (p = 2/3) towards equal allocation:

In[1]:=

Out[1]=

Scope (3) 

Different weighting can be given to the restriction of equal allocation:

In[2]:=

Out[2]=

In[3]:=

Out[3]=

In[4]:=

Out[4]=

Different sequence lengths can be generated:

In[5]:=

Out[5]=

In[6]:=

Out[6]=

In[7]:=

Out[7]=

Users may replace the default symbols for the groups in any way they like:

In[8]:=

Out[8]=

Properties and Relations (2) 

Find the count of all values that follow an "A” when p is near 1/2:

In[9]:=

Out[10]=

Compare to when p is near 1. This shows a strong bias for alternating:

In[11]:=

Out[12]=

Neat Examples (4) 

Complete randomization is achieved by a (p = 1/2) weighting towards selection of the underrepresented group:

In[13]:=

Out[13]=

Complete randomization (p = 1/2) is thus equivalent to simple utilization of RandomChoice, albeit with the sequentially biased algorithm being slower:

In[14]:=

"RandomChoice Runtime: " 0.1218419`"Efron Runtime: " 10.9277715`

Out[16]=

In[17]:=

Out[17]=

In[18]:=

Out[19]=

The variance in allocation percentage for a given group can be diminished by weighting the randomization scheme (p ≠ 1/2):

In[20]:=

Out[21]=

In[22]:=

Out[22]=

In[23]:=

Out[24]=

Explore the options in variance of allocation percentage as a function of the bias p:

In[25]:=

Out[26]=

In[27]:=

Out[27]=