The possibility of determining cosmological parameters on the basis of a wide set of observational data including the Abell-ACO cluster power spectrum and mass function , peculiar velocities of galaxies , the distribution of Ly- \alpha clouds and CMB temperature fluctuations is analyzed .
Using a \chi ^ { 2 } minimization method , assuming \Omega _ { \Lambda } + \Omega _ { matter } = 1 and no contribution from gravity waves , we show that this data set determines quite precisely the values of the spectral index n of the primordial power spectrum , baryon , cold dark matter and massive neutrino density \Omega _ { b } , \Omega _ { cdm } and \Omega _ { \nu } respectively , the Hubble constant h \equiv H _ { 0 } / ( 100 km/s/Mpc ) and the value of the cosmological constant , \Omega _ { \Lambda } .

Varying all parameters , we found that a tilted \Lambda MDM model with one sort of massive neutrinos and the parameters n = 1.12 \pm 0.10 , \Omega _ { m } = 0.41 \pm 0.11 ( \Omega _ { \Lambda } = 0.59 \pm 0.11 ) , \Omega _ { cdm } = 0.31 \pm 0.15 , \Omega _ { \nu } = 0.059 \pm 0.028 , \Omega _ { b } = 0.039 \pm 0.014 and h = 0.70 \pm 0.12 matches observational data best .

\Omega _ { \nu } is higher for more species of massive neutrinos , \sim 0.1 for two and \sim 0.13 for three species . \Omega _ { m } raises by \sim 0.08 and \sim 0.15 respectively .

The 1 \sigma ( 68.3 % ) confidence limits on each cosmological parameter , which are obtained by marginalizing over the other parameters , are 0.82 \leq n \leq 1.39 , 0.19 \leq \Omega _ { m } \leq 1 ( 0 \leq \Omega _ { \Lambda } \leq 0.81 ) , 0 \leq \Omega _ { \nu } \leq 0.17 , 0.021 \leq \Omega _ { b } \leq 0.13 and 0.38 \leq h \leq 0.85 1.5 \leq b _ { cl } \leq 3.5 .
Here b _ { cl } is the cluster bias parameter .
The best-fit parameters for 31 models which are inside of 1 \sigma range of the best model are presented ( Table 4 ) .

Varying only a subset of parameters and fixing the others changes the results .
In particular , if a pure matter model ( \Omega _ { m } = 1 ) is assumed , MDM with \Omega _ { \nu } = 0.22 \pm 0.08 , three species of massive neutrinos and low h = 0.47 \pm 0.05 matches the observational data best .
If a low density Universe \Omega _ { m } = 0.3 is assumed , a \Lambda CDM model without hot dark matter and high h = 0.71 matches the observational data best .
If the primordial power spectrum is scale invariant ( n \equiv 1 ) a low density Universe ( \Omega _ { m } = 0.45 \pm 0.12 ,~ { } ~ { } h = 0.71 \pm 0.13 ) with very little hot dark matter ( \Omega _ { \nu } = 0.04 \pm 0.03 , N _ { \nu } = 1 ) becomes the best fit .

It is shown also that observational data set used here rules out the class of CDM models with h \geq 0.5 , scale invariant primordial power spectrum , zero cosmological constant and spatial curvature at very high confidence level , > 99.99 \% .
The corresponding class of MDM models are ruled out at \sim 95 \% C.L .