rethinkdb_statistical_reql_query_examples.js

##
#Finding the hottest observations per day (or a _single_ maxima for each day)
##
r.db("telemetry_pi").table("observations").create_index
r.db('telemetry_pi').table("observations").group({index: "datetime"}).max("temp").ungroup()("reduction")

##
#Finding some basic statistics and calculating a simple linear regression (y_humidity = alpha + beta_temp * temp)  
##
# https://en.wikipedia.org/wiki/Simple_linear_regression
# https://en.wikipedia.org/wiki/Pearson_product-moment_correlation_coefficient
##
r.object(
  
  "sum_humidity", r.db("telemetry_pi").table("observations").sum("humidity"),
  "sum_temp", r.db("telemetry_pi").table("observations").sum("temp"),
  
  "avg_humidity", r.db("telemetry_pi").table("observations").avg("humidity"),
  "avg_temp", r.db("telemetry_pi").table("observations").avg("temp"),
  
  "sumsq_humidity", r.db('telemetry_pi').table("observations")("humidity").map(function(d){ return d.mul(d); }).sum(), 
  "sumsq_temp", r.db('telemetry_pi').table("observations")("temp").map(function(d){ return d.mul(d); }).sum(),
    
  "sum_products_temp_humidity", 
 		r.map(
      r.db("telemetry_pi").table("observations")("temp"), 
      r.db("telemetry_pi").table("observations")("humidity"),
      function(temp, humidity){ 
      	return temp.mul(humidity); 
    }).sum(),
  
  "n", r.db("telemetry_pi").table("observations").count()
  
).do(function(doc){
  //Calculate the squared averages
  return doc.merge({
    
    beta_hat: doc("n").mul(doc("sum_products_temp_humidity")).sub((doc("sum_humidity").mul(doc("sum_temp")))).div(
    //=========================================================================================
    		doc("n").mul(doc("sumsq_temp")).sub((doc("sum_temp").mul(doc("sum_temp"))))
    ),
    
    "sd_humidity": r.db('telemetry_pi')
      .table("observations")
      ("humidity")
      .map(function(d){
        var x = d.sub(doc("avg_humidity"));
				return x.mul(x); 
    }).avg().do(r.js('(function(x) { return Math.sqrt(x); })')),
    
    "sd_temperature": r.db('telemetry_pi')
      .table("observations")
      ("temp")
      .map(function(d){
        var x = d.sub(doc("avg_temp"));
				return x.mul(x); 
    }).avg().do(r.js('(function(x) { return Math.sqrt(x); })')),
    
    
    "temp_humidity_correlation": doc("sum_products_temp_humidity").div(
      doc("sumsq_temp").mul(doc("sumsq_humidity")).do(r.js('(function(x) { return Math.sqrt(x); })'))
    )
    
  });
}).do(function(doc){
  return doc.merge({
    alpha_hat: (r.expr(1).div(doc("n")).mul(doc("sum_humidity"))).sub(doc("beta_hat").mul(r.expr(1).div(doc("n"))).mul(doc("sum_temp")))      
  });
})