A Quantitative Study on Dream of the Red Chamber: Word-Length Distribution and Authorship Attribution

Table 1

Fitting of extended logarithmic to word-length distribution in six sample texts.

X[i]	Text 20		Text 40		Text 60
	F[i]	NP[i]	F[i]	NP[i]	F[i]	NP[i]
1	1104	1104.00	1084	1084.00	1121	1121.00
2	771	775.60	791	797.05	750	755.69
3	109	99.33	117	98.82	113	100.81
4	16	21.07	8	20.13	16	22.51
	θ = 0.2561, α = 0.4480,		θ = 0.2480, α = 0.4580,		θ = 0.2668, α = 0.4395,
	X^2 = 2.1888,		X^2 = 10.7041,		X^2 = 3.3976,
	DF = 1, C = 0.0011, R^2 = 0.9999		DF = 1, C = 0.0054, R^2 = 0.9994		DF = 1, C = 0.0017, R^2 = 0.9998
X[i]	Text 80		Text 100		Text 120
	F[i]	NP[i]	F[i]	NP[i]	F[i]	NP[i]
1	1027	1027.00	1022	1022.00	962	962.00
2	863	855.51	874	869.27	913	912.72
3	87	98.97	85	92.92	105	105.54
4	22	15.27	19	15.81	20	19.74
5	0	2.65
6	1	0.61
	θ = 0.2314, α = 0.4865,		θ = 0.2138, α = 0.4890,		θ = 0.2313, α = 0.5190,
	X^2 = 6.0484,		X^2 = 1.3447,		X^2 = 0.0062,
	DF = 2, C = 0.0030, R^2 = 0.9998		DF = 1, C = 0.0007, R^2 = 0.9999		DF = 1, C = 0.0000, R^2 = 1.0000

Note. X[i] is the observed classes of word length; F[i] is the observed frequency; NP[i] is the calculated frequency of the extended logarithmic (θ, α) distribution model.